hh.sePublikationer
Ändra sökning
Avgränsa sökresultatet
1234567 151 - 200 av 2367
RefereraExporteraLänk till träfflistan
Permanent länk
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annat format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annat språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
Träffar per sida
  • 5
  • 10
  • 20
  • 50
  • 100
  • 250
Sortering
  • Standard (Relevans)
  • Författare A-Ö
  • Författare Ö-A
  • Titel A-Ö
  • Titel Ö-A
  • Publikationstyp A-Ö
  • Publikationstyp Ö-A
  • Äldst först
  • Nyast först
  • Skapad (Äldst först)
  • Skapad (Nyast först)
  • Senast uppdaterad (Äldst först)
  • Senast uppdaterad (Nyast först)
  • Disputationsdatum (tidigaste först)
  • Disputationsdatum (senaste först)
  • Standard (Relevans)
  • Författare A-Ö
  • Författare Ö-A
  • Titel A-Ö
  • Titel Ö-A
  • Publikationstyp A-Ö
  • Publikationstyp Ö-A
  • Äldst först
  • Nyast först
  • Skapad (Äldst först)
  • Skapad (Nyast först)
  • Senast uppdaterad (Äldst först)
  • Senast uppdaterad (Nyast först)
  • Disputationsdatum (tidigaste först)
  • Disputationsdatum (senaste först)
Markera
Maxantalet träffar du kan exportera från sökgränssnittet är 250. Vid större uttag använd dig av utsökningar.
  • 151.
    Ak, Abdullah Cihan
    et al.
    Istanbul Technical University, Istanbul, Turkey.
    Aksoy, Eren
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Sariel, Sanem
    Istanbul Technical University, Istanbul, Turkey.
    Learning Failure Prevention Skills for Safe Robot Manipulation2023Ingår i: IEEE Robotics and Automation Letters, E-ISSN 2377-3766, Vol. 8, nr 12, s. 7994-8001Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    Robots are more capable of achieving manipulation tasks for everyday activities than before. However, the safety of manipulation skills that robots employ is still an open problem. Considering all possible failures during skill learning increases the complexity of the process and restrains learning an optimal policy. Nonetheless, safety-focused modularity in the acquisition of skills has not been adequately addressed in previous works. For that purpose, we reformulate skills as base and failure prevention skills, where base skills aim at completing tasks and failure prevention skills aim at reducing the risk of failures to occur. Then, we propose a modular and hierarchical method for safe robot manipulation by augmenting base skills by learning failure prevention skills with reinforcement learning and forming a skill library to address different safety risks. Furthermore, a skill selection policy that considers estimated risks is used for the robot to select the best control policy for safe manipulation. Our experiments show that the proposed method achieves the given goal while ensuring safety by preventing failures. We also show that with the proposed method, skill learning is feasible and our safe manipulation tools can be transferred to the real environment © 2023 IEEE

  • 152.
    Holtzberg, Joel
    et al.
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Thomsen, Michel
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Åkesson, Maria
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Learning Management Systems in Flexible Learning Environments - A Study of Teachers’ Experiences2023Ingår i: Design, Learning, and Innovation: 7th EAI International Conference, DLI 2022, Faro, Portugal, November 21-22, 2022, Proceedings / [ed] Eva Brooks; Jeanette Sjöberg; Anders Kalsgaard Møller; Emma Edstrand, Cham: Springer, 2023, Vol. 493, s. 3-21Konferensbidrag (Refereegranskat)
    Abstract [en]

    Digital transformation in education is expected to progress teaching and learning. To meet this expectation, new types of classrooms called flexible learning environments are designed where digital resources such as learning management systems (LMS) are integrated. This raises the question of how LMS are experienced by teachers in flexible learning environments and how their teaching practice and competence development is supported by the LMS. In this study, ten teachers working in flexible working environments have been interviewed about their experiences with LMS. The study resulted in four themes of experiences (1) Lack of adoption, (2) Control within the system, (3) Collaboration and competence development, (4) Direct feedback and interactions. The insights of the study contributes with implications for choosing and integrating LMS in flexible learning environments. © 2023, ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering.

  • 153.
    Jiang, Fan
    et al.
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Ge, Yu
    Chalmers University, Gothenburg, Sweden.
    Zhu, Meifang
    Lund University, Lund, Sweden.
    Wymeersch, Henk
    Chalmers University, Gothenburg, Sweden.
    Tufvesson, Fredrik
    Lund University, Lund, Sweden.
    Low-complexity Channel Estimation and Localization with Random Beamspace Observations2023Ingår i: ICC 2023 - IEEE International Conference on Communications / [ed] Michele Zorzi; Meixia Tao; Walid Saad, Piscataway, NJ: IEEE, 2023, s. 5985-5990Konferensbidrag (Refereegranskat)
    Abstract [en]

    We investigate the problem of low-complexity, high-dimensional channel estimation with beamspace observations, for the purpose of localization. Existing work on beamspace ESPRIT (estimation of signal parameters via rotational invariance technique) approaches requires either a shift-invariance structure of the transformation matrix, or a full-column rank condition. We extend these beamspace ESPRIT methods to a case when neither of these conditions is satisfied, by exploiting the full-row rank of the transformation matrix. We first develop a tensor decomposition-based approach, and further design a matrix-based ESPRIT method to achieve auto-pairing of the channel parameters, with reduced complexity. Numerical simulations show that the proposed methods work in the challenging scenario, and the matrix-based ESPRIT approach achieves better performance than the tensor ESPRIT method. © 2023 IEEE

  • 154.
    de Capretz, Pontus Olsson
    et al.
    Skåne University Hospital, Lund, Sweden; Lund University, Lund, Sweden.
    Björkelund, Anders
    Lund University, Lund, Sweden.
    Björk, Jonas
    Lund University, Lund, Sweden; Skåne University Hospital, Lund, Sweden.
    Ohlsson, Mattias
    Högskolan i Halmstad, Akademin för informationsteknologi. Lund University, Lund, Sweden.
    Mokhtari, Arash
    Skåne University Hospital, Lund, Sweden; Lund University, Lund, Sweden.
    Nyström, Axel
    Lund University, Lund, Sweden.
    Ekelund, Ulf
    Skåne University Hospital, Lund, Sweden; Lund University, Lund, Sweden.
    Machine learning for early prediction of acute myocardial infarction or death in acute chest pain patients using electrocardiogram and blood tests at presentation2023Ingår i: BMC Medical Informatics and Decision Making, E-ISSN 1472-6947, Vol. 23, nr 1, s. 1-10, artikel-id 25Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    Aims: In the present study, we aimed to evaluate the performance of machine learning (ML) models for identification of acute myocardial infarction (AMI) or death within 30 days among emergency department (ED) chest pain patients. Methods and results: Using data from 9519 consecutive ED chest pain patients, we created ML models based on logistic regression or artificial neural networks. Model inputs included sex, age, ECG and the first blood tests at patient presentation: High sensitivity TnT (hs-cTnT), glucose, creatinine, and hemoglobin. For a safe rule-out, the models were adapted to achieve a sensitivity > 99% and a negative predictive value (NPV) > 99.5% for 30-day AMI/death. For rule-in, we set the models to achieve a specificity > 90% and a positive predictive value (PPV) of > 70%. The models were also compared with the 0 h arm of the European Society of Cardiology algorithm (ESC 0 h); An initial hs-cTnT < 5 ng/L for rule-out and ≥ 52 ng/L for rule-in. A convolutional neural network was the best model and identified 55% of the patients for rule-out and 5.3% for rule-in, while maintaining the required sensitivity, specificity, NPV and PPV levels. ESC 0 h failed to reach these performance levels. Discussion: An ML model based on age, sex, ECG and blood tests at ED arrival can identify six out of ten chest pain patients for safe early rule-out or rule-in with no need for serial blood tests. Future studies should attempt to improve these ML models further, e.g. by including additional input data. © 2023, The Author(s).

  • 155.
    Alabdallah, Abdallah
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Machine Learning Survival Models: Performance and Explainability2023Licentiatavhandling, sammanläggning (Övrigt vetenskapligt)
    Abstract [en]

    Survival analysis is an essential statistics and machine learning field in various critical applications like medical research and predictive maintenance. In these domains understanding models' predictions is paramount. While machine learning techniques are increasingly applied to enhance the predictive performance of survival models, they simultaneously sacrifice transparency and explainability. 

    Survival models, in contrast to regular machine learning models, predict functions rather than point estimates like regression and classification models. This creates a challenge regarding explaining such models using the known off-the-shelf machine learning explanation techniques, like Shapley Values, Counterfactual examples, and others.   

    Censoring is also a major issue in survival analysis where the target time variable is not fully observed for all subjects. Moreover, in predictive maintenance settings, recorded events do not always map to actual failures, where some components could be replaced because it is considered faulty or about to fail in the future based on an expert's opinion. Censoring and noisy labels create problems in terms of modeling and evaluation that require to be addressed during the development and evaluation of the survival models.

    Considering the challenges in survival modeling and the differences from regular machine learning models, this thesis aims to bridge this gap by facilitating the use of machine learning explanation methods to produce plausible and actionable explanations for survival models. It also aims to enhance survival modeling and evaluation revealing a better insight into the differences among the compared survival models.

    In this thesis, we propose two methods for explaining survival models which rely on discovering survival patterns in the model's predictions that group the studied subjects into significantly different survival groups. Each pattern reflects a specific survival behavior common to all the subjects in their respective group. We utilize these patterns to explain the predictions of the studied model in two ways. In the first, we employ a classification proxy model that can capture the relationship between the descriptive features of subjects and the learned survival patterns. Explaining such a proxy model using Shapley Values provides insights into the feature attribution of belonging to a specific survival pattern. In the second method, we addressed the "what if?" question by generating plausible and actionable counterfactual examples that would change the predicted pattern of the studied subject. Such counterfactual examples provide insights into actionable changes required to enhance the survivability of subjects.

    We also propose a variational-inference-based generative model for estimating the time-to-event distribution. The model relies on a regression-based loss function with the ability to handle censored cases. It also relies on sampling for estimating the conditional probability of event times. Moreover, we propose a decomposition of the C-index into a weighted harmonic average of two quantities, the concordance among the observed events and the concordance between observed and censored cases. These two quantities, weighted by a factor representing the balance between the two, can reveal differences between survival models previously unseen using only the total Concordance index. This can give insight into the performances of different models and their relation to the characteristics of the studied data.

    Finally, as part of enhancing survival modeling, we propose an algorithm that can correct erroneous event labels in predictive maintenance time-to-event data. we adopt an expectation-maximization-like approach utilizing a genetic algorithm to find better labels that would maximize the survival model's performance. Over iteration, the algorithm builds confidence about events' assignments which improves the search in the following iterations until convergence.

    We performed experiments on real and synthetic data showing that our proposed methods enhance the performance in survival modeling and can reveal the underlying factors contributing to the explainability of survival models' behavior and performance.

    Ladda ner fulltext (pdf)
    fulltext
  • 156.
    Chen, Kunru
    et al.
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Rögnvaldsson, Thorsteinn
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Nowaczyk, Sławomir
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Pashami, Sepideh
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Klang, Jonas
    Toyota Material Handling Manufacturing Sweden AB, Mjölby, Sweden.
    Sternelöv, Gustav
    Toyota Material Handling Manufacturing Sweden AB, Mjölby, Sweden.
    Material handling machine activity recognition by context ensemble with gated recurrent units2023Ingår i: Engineering applications of artificial intelligence, ISSN 0952-1976, E-ISSN 1873-6769, Vol. 126, nr Part C, artikel-id 106992Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    Research on machine activity recognition (MAR) is drawing more attention because MAR can provide productivity monitoring for efficiency optimization, better maintenance scheduling, product design improvement, and potential material savings. A particular challenge of MAR for human-operated machines is the overlap when transiting from one activity to another: during transitions, operators often perform two activities simultaneously, e.g., lifting the fork already while approaching a rack, so the exact time when one activity ends and another begins is uncertain. Machine learning models are often uncertain during such activity transitions, and we propose a novel ensemble-based method adapted to fuzzy transitions in a forklift MAR problem. Unlike traditional ensembles, where models in the ensemble are trained on different subsets of data, or with costs that force them to be diverse in their responses, our approach is to train a single model that predicts several activity labels, each under a different context. These individual predictions are not made by independent networks but are made using a structure that allows for sharing important features, i.e., a context ensemble. The results show that the gated recurrent unit network can provide medium or strong confident context ensembles for 95% of the cases in the test set, and the final forklift MAR result achieves accuracies of 97% for driving and 90% for load-handling activities. This study is the first to highlight the overlapping activity issue in MAR problems and to demonstrate that the recognition results can be significantly improved by designing a machine learning framework that addresses this issue. © 2023 The Author(s)

  • 157.
    Vettoruzzo, Anna
    et al.
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Bouguelia, Mohamed-Rafik
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Rögnvaldsson, Thorsteinn
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Meta-Learning from Multimodal Task Distributions Using Multiple Sets of Meta-Parameters2023Ingår i: 2023 International Joint Conference on Neural Networks (IJCNN), Piscataway, NJ: IEEE, 2023, s. 1-8Konferensbidrag (Refereegranskat)
    Abstract [en]

    Meta-learning or learning to learn involves training a model on various learning tasks in a way that allows it to quickly learn new tasks from the same distribution using only a small amount of training data (i.e., few-shot learning). Current meta-learning methods implicitly assume that the distribution over tasks is unimodal and consists of tasks belonging to a common domain, which significantly reduces the variety of task distributions they can handle. However, in real-world applications, tasks are often very diverse and come from multiple different domains, making it challenging to meta-learn common knowledge shared across the entire task distribution. In this paper, we propose a method for meta-learning from a multimodal task distribution. The proposed method learns multiple sets of meta-parameters (acting as different initializations of a neural network model) and uses a task encoder to select the best initialization to fine-tune for a new task. More specifically, with a few training examples from a task sampled from an unknown mode, the proposed method predicts which set of meta-parameters (i.e., model’s initialization) would lead to a fast adaptation and a good post-adaptation performance on that task. We evaluate the proposed method on a diverse set of few-shot regression and image classification tasks. The results demonstrate the superiority of the proposed method compared to other state of-the-art meta-learning methods and the benefit of learning multiple model initializations when tasks are sampled from a multimodal task distribution. © 2023 IEEE.

    Ladda ner fulltext (pdf)
    Meta-Learning from Multimodal Task Distributions Using Multiple Sets of Meta-Parameters
  • 158.
    Galozy, Alexander
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Mobile Health Interventions through Reinforcement Learning2023Doktorsavhandling, sammanläggning (Övrigt vetenskapligt)
    Abstract [en]

    This thesis presents work conducted in the domain of sequential decision-making in general and Bandit problems in particular, tackling challenges from a practical and theoretical perspective, framed in the contexts of mobile Health. The early stages of this work have been conducted in the context of the project ``improving Medication Adherence through Person-Centred Care and Adaptive Interventions'' (iMedA) which aims to provide personalized adaptive interventions to hypertensive patients, supporting them in managing their medication regimen. The focus lies on inadequate medication adherence (MA), a pervasive issue where patients do not take their medication as instructed by their physician. The selection of individuals for intervention through secondary database analysis on Electronic Health Records (EHRs) was a key challenge and is addressed through in-depth analysis of common adherence measures, development of prediction models for MA, and discussions on limitations of such approaches for analyzing MA. Providing personalized adaptive interventions is framed in several bandit settings and addresses the challenge of delivering relevant interventions in environments where contextual information is unreliable and full of noise. Furthermore, the need for good initial policies is explored and improved in the latent-bandits setting, utilizing prior collected data to optimal selection the best intervention at every decision point. As the final concluding work, this thesis elaborates on the need for privacy and explores different privatization techniques in the form of noise-additive strategies using a realistic recommendation scenario.         

    The contributions of the thesis can be summarised as follows: (1) Highlighting the issues encountered in measuring MA through secondary database analysis and providing recommendations to address these issues, (2) Investigating machine learning models developed using EHRs for MA prediction and extraction of common refilling patterns through EHRs, (3) formal problem definition for a novel contextual bandit setting with context uncertainty commonly encountered in Mobile Health and development of an algorithm designed for such environments. (4) Algorithmic improvements, equipping the agent with information-gathering capabilities for active action selection in the latent bandit setting, and (5) exploring important privacy aspects using a realistic recommender scenario.   

    Ladda ner fulltext (pdf)
    Thesis Fulltext
  • 159.
    Hoveskog, Maya
    et al.
    Högskolan i Halmstad, Akademin för företagande, innovation och hållbarhet.
    Holmén, Magnus
    Högskolan i Halmstad, Akademin för företagande, innovation och hållbarhet.
    Ernest, Anya
    R&D, Connected Experience Innovation, Polestar, Sweden.
    Bergquist, Magnus
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Mobilizing Service Ecosystems for Sustainability – the Case of Polestar2023Ingår i: NBM 2023: Proceedings of the 8th International Conference on New Business Models / [ed] Abel Diaz Gonzalez; Juliette Koning; Nancy Bocken, Maastricht: Maastricht University Press , 2023Konferensbidrag (Refereegranskat)
  • 160.
    Menon, Heera
    et al.
    Lund University, Lund, Sweden.
    Jeddi, Hossein
    Högskolan i Halmstad, Akademin för informationsteknologi. Lund University, Lund, Sweden.
    Morgan, Nicholas Paul
    École Polytechnique Fédérale De Lausanne, Lausanne, Switzerland.
    Fontcuberta i Morral, Anna
    École Polytechnique Fédérale De Lausanne, Lausanne, Switzerland.
    Pettersson, Håkan
    Högskolan i Halmstad, Akademin för informationsteknologi. Lund University, Lund, Sweden.
    Borg, Mattias
    Lund University, Lund, Sweden.
    Monolithic InSb nanostructure photodetectors on Si using rapid melt growth2023Ingår i: Nanoscale Advances, E-ISSN 2516-0230, Vol. 5, nr 4, s. 1152-1162Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    Monolithic integration of InSb on Si could be a key enabler for future electronic and optoelectronic applications. In this work, we report the fabrication of InSb metal-semiconductor-metal photodetectors directly on Si using a CMOS-compatible process known as rapid melt growth. Fourier transform spectroscopy demonstrates a spectrally resolved photocurrent peak from a single crystalline InSb nanostructure with dimensions of 500 nm × 1.1 μm × 120 nm. Time-dependent optical characterization of a device under 1550 nm illumination indicated a stable photoresponse with responsivity of 0.50 A W−1 at 16 nW illumination, with a time constant in the range of milliseconds. Electron backscatter diffraction spectroscopy revealed that the single crystalline InSb nanostructures contain occasional twin defects and crystal lattice twist around the growth axis, in addition to residual strain, possibly causing the observation of a low-energy tail in the detector response extending the photosensitivity out to 10 μm wavelengths (0.12 eV) at 77 K. © 2023 RSC.

  • 161.
    Ding, Yijie
    et al.
    University of Electronic Science and Technology of China, Quzhou, China.
    Tiwari, Prayag
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Guo, Fei
    Central South University, Changsha, China.
    Zou, Quan
    University of Electronic Science and Technology of China, Chengdu, China; University of Electronic Science and Technology of China, Chengdu, China.
    Multi-correntropy fusion based fuzzy system for predicting DNA N4-methylcytosine sites2023Ingår i: Information Fusion, ISSN 1566-2535, E-ISSN 1872-6305, Vol. 100, s. 1-10, artikel-id 101911Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    The identification of DNA N4-methylcytosine (4mC) sites is an important field of bioinformatics. Statistical learning methods and deep learning have been applied in this direction. The previous methods focused on feature representation and feature selection, and did not take into account the deviation of noise samples for recognition. Moreover, these models were not established from the perspective of prediction error distribution. To solve the problem of complex error distribution, we propose a maximum multi-correntropy criterion based kernelized higher-order fuzzy inference system (MMC-KHFIS), which is constructed with multi-correntropy fusion. There are 6 4mC and 8 UCI data sets are employed to evaluate our model. The MMC-KHFIS achieves better performance in the experiment. © 2023

  • 162.
    Taghiyarrenani, Zahra
    et al.
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Nowaczyk, Sławomir
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Pashami, Sepideh
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Bouguelia, Mohamed-Rafik
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Multi-Domain Adaptation for Regression under Conditional Distribution Shift2023Ingår i: Expert systems with applications, ISSN 0957-4174, E-ISSN 1873-6793, Vol. 224, artikel-id 119907Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    Domain adaptation (DA) methods facilitate cross-domain learning by minimizing the marginal or conditional distribution shift between domains. However, the conditional distribution shift is not well addressed by existing DA techniques for the cross-domain regression learning task. In this paper, we propose Multi-Domain Adaptation for Regression under Conditional shift (DARC) method. DARC constructs a shared feature space such that linear regression on top of that space generalizes to all domains. In other words, DARC aligns different domains and makes explicit the task-related information encoded in the values of the dependent variable. It is achieved using a novel Pairwise Similarity Preserver (PSP) loss function. PSP incentivizes the differences between the outcomes of any two samples, regardless of their domain(s), to match the distance between these samples in the constructed space.

    We perform experiments in both two-domain and multi-domain settings. The two-domain setting is helpful, especially when one domain contains few available labeled samples and can benefit from adaptation to a domain with many labeled samples. The multi-domain setting allows several domains, each with limited data, to be adapted collectively; thus, multiple domains compensate for each other’s lack of data. The results from all the experiments conducted both on synthetic and real-world datasets confirm the effectiveness of DARC. © 2023 The Authors

  • 163.
    Chalangar, Ebrahim
    et al.
    Högskolan i Halmstad. Linköping University, Linkoping, Sweden.
    Mustafa, Elfatih
    Linköping University, Linkoping, Sweden.
    Nur, Omer
    Linköping University, Linkoping, Sweden.
    Willander, Magnus
    Linköping University, Linkoping, Sweden.
    Pettersson, Håkan
    Högskolan i Halmstad, Akademin för informationsteknologi. Linköping University, Linkoping, Sweden; Nanolund, Lund, Sweden.
    Nanopatterned rGO/ZnO: Al seed layer for vertical growth of single ZnO nanorods2023Ingår i: Nanotechnology, ISSN 0957-4484, E-ISSN 1361-6528, Vol. 34, nr 25, s. 1-7, artikel-id 255301Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    In this work, we demonstrate a novel low-cost template-assisted route to synthesize vertical ZnO nanorod arrays on Si (100). The nanorods were grown on a patterned double seed layer comprised of reduced graphene oxide (rGO) and Al-doped ZnO nanoparticles. The seed layer was fabricated by spray-coating the substrate with graphene and then dip-coating it into a Al-doped ZnO sol-gel solution. The growth template was fabricated from a double-layer resist, spin-coated on top of the rGO/ZnO:Al seed layer, and patterned by colloidal lithography. The results show a successful chemical bath deposition of vertically aligned ZnO nanorods with controllable diameter and density in the nanoholes in the patterned resist mask. Our novel method can presumably be used to fabricate electronic devices on virtually any smooth substrate with a thermal budget of 1 min at 300 °C with the seed layer acting as a conductive strain-relieving back contact. The top contact can simply be made by depositing a suitable transparent conductive oxide or metal, depending on the specific application. © 2023 The Author(s). Published by IOP Publishing Ltd.

  • 164.
    Cooney, Martin
    et al.
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Sjöberg, Jeanette
    Högskolan i Halmstad, Akademin för lärande, humaniora och samhälle.
    Navigating the Current “New World” of Teaching with Technology: A Glimpse into Our Teachers’ Minds2023Ingår i: Design, Learning, and Innovation: 7th EAI International Conference, DLI 2022, Faro, Portugal, November 21–22, 2022, Proceedings / [ed] Eva Brooks; Jeanette Sjöberg; Anders Kalsgaard Møller; Emma Edstrand, Cham: Springer, 2023, s. 135-152Konferensbidrag (Refereegranskat)
    Abstract [en]

    The COVID-19 pandemic helped spark a surge in innovative usages of technology in education, from robot-based remote graduation ceremonies to immersive learning through extended reality, meetings in fantastical game worlds, automatic examination methods, and flexible learning options such as hybrid classes. It’s been said that we can’t go back to “normal” because this is normal now–but what exactly is today’s “new normal”? The current paper reports on the results of an anonymous online survey conducted with 42 teachers in business, IT, nursing, and education at our university in October 2021, to gain insight into where some teachers on the “front lines” currently stand on the use of technology in education. Some insights included that: More teachers than we had expected were using robotics and extended reality (XR), suggesting that silo effects can exist in education, even at small universities; furthermore, the rates of teachers who had seen such usage seemed close to the rates of teachers who had tried using them, suggesting the usefulness of raising awareness to promote professional digital competence (PDC). Rates for using games and exam tools were lower than expected, despite the availability of game platforms and a growing need to consider the threat of how technology can be misused to cheat in exams, possibly due to teachers’ limited time for pedagogical development. Also, teachers appeared to have strong and differing opinions about learning formats, although a general preference was observed for physical classes and exams, and hybrid teacher meetings. Our aim is that these results will be used by our university’s pedagogical center to support our teachers’ PDC and uses of edtech in the near future. © 2023, ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering.

  • 165.
    Hjärtström, Malin
    et al.
    Lund University, Lund, Sweden.
    Dihge, Looket
    Lund University, Lund, Sweden; Skåne University Hospital, Malmö, Sweden.
    Bendahl, Pär-Ola
    Lund University, Lund, Sweden.
    Skarping, Ida
    Lund University, Lund, Sweden; Skåne University Hospital, Malmö, Sweden.
    Ellbrant, Julia
    Lund University, Lund, Sweden; Skåne University Hospital, Malmö, Sweden.
    Ohlsson, Mattias
    Högskolan i Halmstad, Akademin för informationsteknologi. Skåne University Hospital, Malmö, Sweden.
    Rydén, Lisa
    Lund University, Lund, Sweden; Skåne University Hospital, Malmö, Sweden.
    Noninvasive Staging of Lymph Node Status in Breast Cancer Using Machine Learning: External Validation and Further Model Development2023Ingår i: JMIR Cancer, E-ISSN 2369-1999, Vol. 9, artikel-id e46474Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    Background: Most patients diagnosed with breast cancer present with a node-negative disease. Sentinel lymph node biopsy (SLNB) is routinely used for axillary staging, leaving patients with healthy axillary lymph nodes without therapeutic effects but at risk of morbidities from the intervention. Numerous studies have developed nodal status prediction models for noninvasive axillary staging using postoperative data or imaging features that are not part of the diagnostic workup. Lymphovascular invasion (LVI) is a top-ranked predictor of nodal metastasis; however, its preoperative assessment is challenging.Objective: This paper aimed to externally validate a multilayer perceptron (MLP) model for noninvasive lymph node staging (NILS) in a large population-based cohort (n=18,633) and develop a new MLP in the same cohort. Data were extracted from the Swedish National Quality Register for Breast Cancer (NKBC, 2014-2017), comprising only routinely and preoperatively available documented clinicopathological variables. A secondary aim was to develop and validate an LVI MLP for imputation of missing LVI status to increase the preoperative feasibility of the original NILS model. Methods: Three nonoverlapping cohorts were used for model development and validation. A total of 4 MLPs for nodal status and 1 LVI MLP were developed using 11 to 12 routinely available predictors. Three nodal status models were used to account for the different availabilities of LVI status in the cohorts and external validation in NKBC. The fourth nodal status model was developed for 80% (14,906/18,663) of NKBC cases and validated in the remaining 20% (3727/18,663). Three alternatives for imputation of LVI status were compared. The discriminatory capacity was evaluated using the validation area under the receiver operating characteristics curve (AUC) in 3 of the nodal status models. The clinical feasibility of the models was evaluated using calibration and decision curve analyses.Results: External validation of the original NILS model was performed in NKBC (AUC 0.699, 95% CI 0.690-0.708) with good calibration and the potential of sparing 16% of patients with node-negative disease from SLNB. The LVI model was externally validated (AUC 0.747, 95% CI 0.694-0.799) with good calibration but did not improve the discriminatory performance of the nodal status models. A new nodal status model was developed in NKBC without information on LVI (AUC 0.709, 95% CI: 0.688-0.729), with excellent calibration in the holdout internal validation cohort, resulting in the potential omission of 24% of patients from unnecessary SLNBs.Conclusions: The NILS model was externally validated in NKBC, where the imputation of LVI status did not improve the model's discriminatory performance. A new nodal status model demonstrated the feasibility of using register data comprising only the variables available in the preoperative setting for NILS using machine learning. Future steps include ongoing preoperative validation of the NILS model and extending the model with, for example, mammography images. © Malin Hjärtström, Looket Dihge, Pär-Ola Bendahl, Ida Skarping, Julia Ellbrant, Mattias Ohlsson, Lisa Rydén.

  • 166.
    Thunberg, Johan
    et al.
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Bernard, Florian
    University of Bonn, Bonn, Germany.
    Non-negative Spherical Relaxations for Universe-Free Multi-matching and Clustering2023Ingår i: Image Analysis: 22nd Scandinavian Conference, SCIA 2023, Sirkka, Finland, April 18–21, 2023, Proceedings, Part II / [ed] Gade, Rikke; Felsberg, Michael; Kämäräinen, Joni-Kristian, Cham: Springer, 2023, Vol. 13886, s. 260-277Konferensbidrag (Refereegranskat)
    Abstract [en]

    We propose a novel non-negative spherical relaxation for optimization problems over binary matrices with injectivity constraints, which in particular has applications in multi-matching and clustering. We relax respective binary matrix constraints to the (high-dimensional) non-negative sphere. To optimize our relaxed problem, we use a conditional power iteration method to iteratively improve the objective function, while at same time sweeping over a continuous scalar parameter that is (indirectly) related to the universe size (or number of clusters). Opposed to existing procedures that require to fix the integer universe size before optimization, our method automatically adjusts the analogous continuous parameter. Furthermore, while our approach shares similarities with spectral multi-matching and spectral clustering, our formulation has the strong advantage that we do not rely on additional post-processing procedures to obtain binary results. Our method shows compelling results in various multi-matching and clustering settings, even when compared to methods that use the ground truth universe size (or number of clusters). © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

  • 167.
    Aslam, Muhammad Shamrooz
    et al.
    School of Automation, Guangxi University of Science and Technology, Liuzhou, P.R. China.
    Tiwari, Prayag
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Pandey, Hari Mohan
    Data Science and Artificial Intelligence, Department of Information and Computing, Bournemouth University, Bournemouth, United Kingdom.
    Band, Shahab S.
    Future Technology Research Center, College of Future, National Yunlin University of Science and Technology, Douliou, Taiwan, Republic of China.
    Observer–Based Control for a New Stochastic Maximum Power Point tracking for Photovoltaic Systems With Networked Control System2023Ingår i: IEEE transactions on fuzzy systems, ISSN 1063-6706, E-ISSN 1941-0034, Vol. 31, nr 6, s. 1870-1884Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    This study discusses the new stochastic maximum power point tracking (MPPT) control approach towards the photovoltaic cells (PCs). PC generator is isolated from the grid, resulting in a direct current (DC) microgrid that can provide changing loads. In the course of the nonlinear systems through the time-varying delays, we proposed a Networked Control Systems (NCSs) beneath an event-triggered approach basically in the fuzzy system. In this scenario, we look at how random, variable loads impact the PC generator's stability and efficiency. The basic premise of this article is to load changes and the value matching to a Markov chain. PC generators are complicated nonlinear systems that pose a modeling problem. Transforming this nonlinear PC generator model into the Takagi–Sugeno (T–S) fuzzy model is another option. Takagi–Sugeno (T–S) fuzzy model is presented in a unified framework, for which 1) the fuzzy observer–based on this premise variables can be used for approximately in the infinite states to the present system, 2) the fuzzy observer–based controller can be created using this same premises be the observer, and 3) to reduce the impact of transmission burden, an event-triggered method can be investigated. Simulating in the PC generator model for the realtime climate data obtained in China demonstrates the importance of our method. In addition, by using a new Lyapunov–Krasovskii functional (LKF) for combining to the allowed weighting matrices incorporating mode-dependent integral terms, the developed model can be stochastically stable and achieves the required performances. Based on the T-P transformation, a new depiction of the nonlinear system is derived in two separate steps in which an adequate controller input is guaranteed in the first step and an adequate vertex polytope is ensured in the second step. To present the potential of our proposed method, we simulate it for PC generators. © 2022 IEEE.

    Ladda ner fulltext (pdf)
    fulltext
  • 168.
    Segata, Michele
    et al.
    University Of Trento, Trento, Italy.
    Casari, Paolo
    University Of Trento, Trento, Italy.
    Lestas, Marios
    Frederick University, Nicosia, Cyprus.
    Tyrovolas, Dimitrios
    Aristotle University Of Thessaloniki, Thessaloniki, Greece.
    Taqua, Saeed
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Karagiannidis, George
    Aristotle University Of Thessaloniki, Thessaloniki, Greece; Lebanese American University, Beirut, Lebanon.
    Liaskos, Christos
    University Of Ioannina, Ioannina, Greece.
    On the Feasibility of RIS-enabled Cooperative Driving2023Ingår i: IEEE Vehicular Networking Conference, VNC / [ed] S. Coleri; O. Altintas; F. Kargl; T. Higuchi; M. Segata; F. Klingler, New York: IEEE, 2023, s. 143-150Konferensbidrag (Refereegranskat)
    Abstract [en]

    Future cooperative autonomous vehicles will require high-performance communication means to support functions such as cooperative maneuvering and cooperative perception. The high-bandwidth requirements of these functions can be met through mmWave communications, whose utilization is often hindered by the harsh propagation conditions of typical vehicular environments. A solution to this problem is the use of reconfigurable intelligent surfaces (RISs), which enable the reflection of signals in a configurable direction, and have recently gained attention in the vehicular domain. In this paper, we provide an initial feasibility study, highlighting the challenges ahead and the performance RISs need to deliver in order to enable this type of communications. Specifically, we utilize CoopeRIS, a simulation framework for RISs integrated into the Plexe/Veins/SUMO ecosystem that we develop as further contribution and will release to the public. © 2023 IEEE.

  • 169.
    Hernandez-Diaz, Kevin
    et al.
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Alonso-Fernandez, Fernando
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Bigun, Josef
    Högskolan i Halmstad, Akademin för informationsteknologi.
    One-Shot Learning for Periocular Recognition: Exploring the Effect of Domain Adaptation and Data Bias on Deep Representations2023Ingår i: IEEE Access, E-ISSN 2169-3536, Vol. 11, s. 100396-100413Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    One weakness of machine-learning algorithms is the need to train the models for a new task. This presents a specific challenge for biometric recognition due to the dynamic nature of databases and, in some instances, the reliance on subject collaboration for data collection. In this paper, we investigate the behavior of deep representations in widely used CNN models under extreme data scarcity for One-Shot periocular recognition, a biometric recognition task. We analyze the outputs of CNN layers as identity-representing feature vectors. We examine the impact of Domain Adaptation on the network layers&#x2019; output for unseen data and evaluate the method&#x2019;s robustness concerning data normalization and generalization of the best-performing layer. We improved state-of-the-art results that made use of networks trained with biometric datasets with millions of images and fine-tuned for the target periocular dataset by utilizing out-of-the-box CNNs trained for the ImageNet Recognition Challenge and standard computer vision algorithms. For example, for the Cross-Eyed dataset, we could reduce the EER by 67% and 79% (from 1.70%and 3.41% to 0.56% and 0.71%) in the Close-World and Open-World protocols, respectively, for the periocular case. We also demonstrate that traditional algorithms like SIFT can outperform CNNs in situations with limited data or scenarios where the network has not been trained with the test classes like the Open-World mode. SIFT alone was able to reduce the EER by 64% and 71.6% (from 1.7% and 3.41% to 0.6% and 0.97%) for Cross-Eyed in the Close-World and Open-World protocols, respectively, and a reduction of 4.6% (from 3.94% to 3.76%) in the PolyU database for the Open-World and single biometric case.

    Ladda ner fulltext (pdf)
    Fulltext
  • 170.
    Mansoor, Ali
    et al.
    Islamic Azad University, Tehran, Iran.
    Fazeli, Mahdi
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Masoud Rahmani, Amir
    Islamic Azad University, Tehran, Iran.
    Reshadi, Midia
    Islamic Azad University, Tehran, Iran.
    Optimized reverse converters with multibit soft error correction support at 7nm technology2023Ingår i: Computers & electrical engineering, ISSN 0045-7906, E-ISSN 1879-0755, Vol. 107, artikel-id 108654Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    Residue number system (RNS) speeds up digital signal processing systems involving dominant addition and multiplication. Addition and multiplication are accelerated further and performed with a balanced performance on one-hot coded (OHC) residue digits. However, the high complexity of the RNS reverse converter (RC) may kill the performance gain. This paper proposes a high-speed and scalable RNS-RC for both regular and one-hot RNS. For redundant RNS (RRNS), an RRNS-RC is proposed, which based on majority-voting between OHC residue digits, corrects multibit soft errors occurring in a single residue channel. With pass-transistor logic and low-power FinFETs, the proposed RCs are optimized. The simulated RRNS-RC corrected 98.5% of soft errors, while consuming 7, 6.2 and 0.4% of the system's area, leakage power and dynamic power, respectively. As compared to the leading lookup table RC in the literature, RNS-RC exhibited 10.3X, 4.4X and 1.8X savings in area, average power, and delay, respectively. © 2023 The Author(s)

  • 171.
    Zhao, Xinhui
    et al.
    Zhuhai Technician College, Zhuhai, China.
    Liang, Guojun
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Optimizing electric vehicle charging schedules and energy management in smart grids using an integrated GA-GRU-RL approach2023Ingår i: Frontiers in Energy Research, E-ISSN 2296-598X, Vol. 11, artikel-id 1268513Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    Introduction: Smart grid technology is a crucial direction for the future development of power systems, with electric vehicles, especially new energy vehicles, serving as important carriers for smart grids. However, the main challenge faced by smart grids is the efficient scheduling of electric vehicle charging and effective energy management within the grid.

    Methods: To address this issue, we propose a novel approach for intelligent grid electric vehicle charging scheduling and energy management, integrating three powerful technologies: Genetic Algorithm (GA), Gated Recurrent Unit (GRU) neural network, and Reinforcement Learning (RL) algorithm. This integrated approach enables global search, sequence prediction, and intelligent decision-making to optimize electric vehicle charging scheduling and energy management. Firstly, the Genetic Algorithm optimizes electric vehicle charging demands while minimizing peak grid loads. Secondly, the GRU model accurately predicts electric vehicle charging demands and grid load conditions, facilitating the optimization of electric vehicle charging schedules. Lastly, the Reinforcement Learning algorithm focuses on energy management, aiming to minimize grid energy costs while meeting electric vehicle charging demands.

    Results and discussion: Experimental results demonstrate that the method achieves prediction accuracy and recall rates of 97.56% and 95.17%, respectively, with parameters (M) and triggers (G) at 210.04 M and 115.65G, significantly outperforming traditional models. The approach significantly reduces peak grid loads and energy costs while ensuring the fulfilment of electric vehicle charging demands and promoting the adoption of green energy in smart city environments. Copyright © 2023 Zhao and Liang.

  • 172.
    Gharehbaghi, Arash
    et al.
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Partovi, Elaheh
    Amirkabir University of Technology, Tehran, Iran.
    Babic, Ankica
    Linköping University, Linköping, Sweden; University of Bergen, Bergen, Norway.
    Parralel Recurrent Convolutional Neural Network for Abnormal Heart Sound Classification2023Ingår i: Caring is sharing - exploiting the value in data for health and innovation: [33rd Medical Informatics Europe Conference, MIE2023, held in Gothenburg, Sweden, from 22 to 25 May, Amsterdam: IOS Press, 2023, Vol. 302, s. 526-530Konferensbidrag (Refereegranskat)
    Abstract [en]

    This paper presents the results of a study performed on Parallel Convolutional Neural Network (PCNN) toward detecting heart abnormalities from the heart sound signals. The PCNN preserves dynamic contents of the signal in a parallel combination of the recurrent neural network and a Convolutional Neural Network (CNN). The performance of the PCNN is evaluated and compared to the one obtained from a Serial form of the Convolutional Neural Network (SCNN) as well as two other baseline studies: a Long- and Short-Term Memory (LSTM) neural network and a Conventional CNN (CCNN). We employed a well-known public dataset of heart sound signals: the Physionet heart sound. The accuracy of the PCNN, was estimated to be 87.2% which outperforms the rest of the three methods: the SCNN, the LSTM, and the CCNN by 12%, 7%, and 0.5%, respectively. The resulting method can be easily implemented in an Internet of Things platform to be employed as a decision support system for the screening heart abnormalities.

  • 173.
    Friel, R. J.
    et al.
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Norfolk, M.
    Fabrisonic LLC, Columbus, OH, United States.
    Power ultrasonics for additive and hybrid manufacturing2023Ingår i: Power Ultrasonics: Applications of High-Intensity Ultrasound, Second Edition / [ed] Gallego-Juárez, Juan A.; Graff, Karl F.; Lucas, Margaret, Oxford: Woodhead Publishing Limited, 2023, 2, s. 227-242Kapitel i bok, del av antologi (Refereegranskat)
    Abstract [en]

    This chapter explores the ultrasonic additive manufacturing (UAM) process. This process is an advanced, solid-state, metal additive/subtractive hybrid manufacturing process. The process combines high-power ultrasonic welding and computer numerical control milling to fabricate solid metal components, layer by layer, from metal foils. The chapter discusses the process fundamentals and three key abilities of UAM: complicated geometries, dissimilar material bonding, and object embedment. Combining these three key abilities positions UAM as a unique and attractive method to create metal matrix–based freeform multifunctional structures for high-value engineering applications. © 2023 Elsevier Ltd. All rights reserved.

  • 174.
    Nilsson, Felix
    et al.
    HMS Labs, HMS Industrial Networks AB, Halmstad, Sweden.
    Bouguelia, Mohamed-Rafik
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Rögnvaldsson, Thorsteinn
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Practical Joint Human-Machine Exploration of Industrial Time Series Using the Matrix Profile2023Ingår i: Data mining and knowledge discovery, ISSN 1384-5810, E-ISSN 1573-756X, Vol. 37, s. 1-38Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    Technological advancements and widespread adaptation of new technology in industry have made industrial time series data more available than ever before. With this development grows the need for versatile methods for mining industrial time series data. This paper introduces a practical approach for joint human-machine exploration of industrial time series data using the Matrix Profile (MP), and presents some challenges involved. The approach is demonstrated on three real-life industrial data sets to show how it enables the user to quickly extract semantic information, detect cycles, find deviating patterns, and gain a deeper understanding of the time series. A benchmark test is also presented on ECG (electrocardiogram) data, showing that the approach works well in comparison to previously suggested methods for extracting relevant time series motifs. © 2022, The Author(s).

    Ladda ner fulltext (pdf)
    fulltext
  • 175.
    Paluch, Richard
    et al.
    University of Siegen, Siegen, Germany.
    Cerna, Katerina
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Kirschsieper, Dennis
    University of Siegen, Siegen, Germany.
    Müller, Claudia
    University of Siegen, Siegen, Germany.
    Practices of Care in Participatory Design With Older Adults During the COVID-19 Pandemic: Digitally Mediated Study2023Ingår i: Journal of Medical Internet Research, E-ISSN 1438-8871, Vol. 25, artikel-id e45750Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    BACKGROUND: Participatory Design (PD), albeit an established approach in User-Centered Design, comes with specific challenges when working with older adults as research participants. Addressing these challenges relates to the reflection and negotiation of the positionalities of the researchers and research participants and includes various acts of giving and receiving help. During the COVID-19 pandemic, facets of positionalities and (mutual) care became particularly evident in qualitative and participatory research settings. OBJECTIVE: The aim of this paper was to systematically analyze care practices of participatory (design) research, which are to different extents practices of the latter. Using a multiyear PD project with older people that had to take place remotely over many months, we specify different practices of care; how they relate to collaborative work in the design project; and represent foundational practices for sustainable, long-term co-design. Our research questions were "How can digitally-mediated PD work during COVID-19 and can we understand such digital PD as 'care'?" METHODS: Our data comes from the Joint Programming Initiative "More Years, Better Lives" (JPI MYBL), a European Union project that aims to promote digital literacy and technology appropriation among older adults in domestic settings. It targeted the cocreation, by older adults and university researchers, of a mobile demo kit website with cocreated resources, aimed at improving the understanding of use options of digital tools. Through a series of workshops, a range of current IT products was explored by a group of 21 older adults, which served as the basis for joint cocreative work on generating design ideas and prototypes. We reflect on the PD process and examine how the actors enact and manifest care. RESULTS: The use of digital technology allowed the participatory project to continue during the COVID-19 pandemic and accentuated the digital skills of older adults and the improvement of digital literacy as part of "care." We provide empirically based evidence of PD with older adults developing digital literacy and sensitizing concepts, based on the notion of care by Tronto for differentiating aspects and processes of care. The data suggest that it is not enough to focus solely on the technologies and how they are used; it is also necessary to focus on the social structures in which help is available and in which technologies offer opportunities to do care work. CONCLUSIONS: We document that the cocreation of different digital media tools can be used to provide a community with mutual care. Our study demonstrates how research participants effectively enact different forms of care and how such "care" is a necessary basis for a genuinely participatory approach, which became especially meaningful as a form of support during COVID-19. We reflect on how notions of "care" and "caring" that were central to the pandemic response are also central to PD. ©Richard Paluch, Katerina Cerna, Dennis Kirschsieper, Claudia Müller. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 17.07.2023.

  • 176.
    Munir, Sundas
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Pre-deployment Analysis of Smart Contracts2023Licentiatavhandling, sammanläggning (Övrigt vetenskapligt)
    Abstract [en]

    Smart contracts are programs that reside and execute on top of blockchains. These programs commonly perform financial transactions and contain the backend logic of several blockchain-supported applications. The presence of errors and bugs in smart contracts poses security threats to the applications they support. This is especially concerning because operations performed by smart contracts are irreversible after deployment due to the immutable nature of blockchains. Thus, ensuring their correctness and security before deployment is important. For this purpose, several program analysis and verification approaches are being actively researched and applied to smart contracts.

    The volume of research in this area makes it challenging to articulate the state-of-the-art. The first contribution of this thesis is to investigate how predeployment analysis techniques ensure the correctness and security of smart contracts. This investigation factors out the relationship between vulnerabilities in smart contracts and pre-deployment analysis techniques through properties they address.

    Among the range of issues uncovered by the investigation, one notable set pertains to non-deterministic factors involved in the context of contract execution. For example, transactions (function invocations) dispatched to smart contracts are scheduled in non-deterministic order, and asynchronous calls to external services (known as oracles) return in a non-deterministic order. Consequently, these factors may cause data races and non-deterministic bugs in smart contracts. The second contribution of this thesis is to address such issues by unraveling specific forms of data races in Ethereum smart contracts, denoted as transactional data races. The thesis also presents a static analysis approach to detect issues arising from transactional data races.

    In addition, this thesis makes a third contribution relating to a design approach for Domain Specific Languages (DSLs). Research on DSL design approaches has the potential to complement the research on smart contracts, as smart contracts are commonly written using DSLs. This thesis proposes an agile approach for designing a DSL for automotive safety test grounds. This approach enables increased communication and learning between different stakeholders involved in DSL development.

    Finally, this thesis highlights our future research endeavors concerning various forms of concurrency and non-determinism-related issues in smart contracts.

    Ladda ner fulltext (pdf)
    Pre-deployment Analysis of Smart Contracts
  • 177.
    Khan, Taha
    et al.
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Dougherty, Mark
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Predicting mental illness at workplace using machine learning2023Ingår i: Mehran University Research Journal of Engineering and Technology, ISSN 0254-7821, E-ISSN 2413-7219, Vol. 42, nr 1, s. 95-108Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    Mental illness (MI) is a leading cause of workplace absenteeism that often goes unrecognized and untreated. This paper presents a machine learning algorithm for predicting MI at workplace. The dataset consisted of responses from 1259 subjects collected through an online survey using a self-assessed questionnaire on the workplace environment. The responses were used as features for training a support vector machine to predict MI. Statistical analysis using the Guttmann correlation and the analysis of variance was done to determine feature significance. Results using 10-fold cross-validation showed that the model predicted MI with good accuracy. Findings support the feasibility of this approach for MI monitoring at the workplace as it offers an advantage over other technologies e.g., MRI scans, and EEG analysis, previously developed for the objective assessment of MI. © Mehran University of Engineering and Technology 2023

  • 178.
    Khoshkangini, Reza
    et al.
    Department of Computer Science and Media Technology, Malmö University, Malmö, Sweden.
    Sheikholharam Mashhadi, Peyman
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Tegnered, Daniel
    Volvo Group Connected Solutions, Gothenburg, Sweden.
    Lundström, Jens
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Rögnvaldsson, Thorsteinn
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Predicting Vehicle Behavior Using Multi-task Ensemble Learning2023Ingår i: Expert systems with applications, ISSN 0957-4174, E-ISSN 1873-6793, Vol. 212, artikel-id 118716Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    Vehicle utilization analysis is an essential tool for manufacturers to understand customer needs, improve equipment uptime, and to collect information for future vehicle and service development. Typically today, this behavioral modeling is done on high-resolution time-resolved data with features such as GPS position and fuel consumption. However, high-resolution data is costly to transfer and sensitive from a privacy perspective. Therefore, such data is typically only collected when the customer pays for extra services relying on that data. This motivated us to develop a multi-task ensemble approach to transfer knowledge from the high-resolution data and enable vehicle behavior prediction from low-resolution but high dimensional data that is aggregated over time in the vehicles.

    This study proposes a multi-task snapshot-stacked ensemble (MTSSE) deep neural network for vehicle behavior prediction by considering vehicles’ low-resolution operational life records. The multi-task ensemble approach utilizes the measurements to map the low-frequency vehicle usage to the vehicle behaviors defined from the high-resolution time-resolved data. Two data sources are integrated and used: high-resolution data called Dynafleet, and low-resolution so-called Logged Vehicle Data (LVD). The experimental results demonstrate the proposed approach’s effectiveness in predicting the vehicle behavior from low frequency data. With the suggested multi-task snapshot-stacked ensemble deep network, it is shown how low-resolution sensor data can highly contribute to predicting multiple vehicle behaviors simultaneously while using only one single training process. © 2022 The Author(s)

    Ladda ner fulltext (pdf)
    fulltext
  • 179.
    Qu, Zhiguo
    et al.
    Nanjing University of Information Science and Technology, Nanjing, China.
    Tang, Yang
    Nanjing University of Information Science and Technology, Nanjing, China.
    Muhammad, Ghulam
    King Saud University, Riyadh, Saudi Arabia.
    Tiwari, Prayag
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Privacy protection in intelligent vehicle networking: A novel federated learning algorithm based on information fusion2023Ingår i: Information Fusion, ISSN 1566-2535, E-ISSN 1872-6305, Vol. 98, artikel-id 101824Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    Federated learning is an effective technique to solve the problem of information fusion and information sharing in intelligent vehicle networking. However, most of the existing federated learning algorithms generally have the risk of privacy leakage. To address this security risk, this paper proposes a novel personalized federated learning with privacy preservation (PDP-PFL) algorithm based on information fusion. In the first stage of its execution, the new algorithm achieves personalized privacy protection by grading users’ privacy based on their privacy preferences and adding noise that satisfies their privacy preferences. In the second stage of its execution, PDP-PFL performs collaborative training of deep models among different in-vehicle terminals for personalized learning, using a lightweight dynamic convolutional network architecture without sharing the local data of each terminal. Instead of sharing all the parameters of the model as in standard federated learning, PDP-PFL keeps the last layer local, thus adding another layer of data confidentiality and making it difficult for the adversary to infer the image of the target vehicle terminal. It trains a personalized model for each vehicle terminal by “local fine-tuning”. Based on experiments, it is shown that the accuracy of the proposed new algorithm for PDP-PFL calculation can be comparable to or better than that of the FedAvg algorithm and the FedBN algorithm, while further enhancing the protection of data privacy. © 2023 Elsevier B.V.

  • 180.
    Sjöberg, Jeanette
    et al.
    Högskolan i Halmstad, Akademin för lärande, humaniora och samhälle.
    Byttner, Stefan
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Wärnestål, Pontus
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Burgos, Jonathan
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Holmén, Magnus
    Högskolan i Halmstad, Akademin för företagande, innovation och hållbarhet.
    Promoting Life-Long Learning Through Flexible Educational Format for Professionals Within AI, Design and Innovation Management2023Ingår i: Design, Learning, and Innovation: 7th EAI International Conference, DLI 2022, Faro, Portugal, November 21–22, 2022, Proceedings / [ed] Eva Brooks; Jeanette Sjöberg; Anders Kalsgaard Møller; Emma Edstrand, Cham: Springer, 2023, s. 38-47Konferensbidrag (Refereegranskat)
    Abstract [en]

    In recent years, the concept of lifelong learning has been emphasized in relation to higher education, with a bearing idea of the possibility for the individual for a continuous, self-motivated pursuit of gaining knowledge for both personal and professional reasons, provided by higher education institutions (HEI:s). But how can this actually be done in practice? In this paper we present an ongoing project called MAISTR, which is a collaboration between Swedish HEI:s and industry with the aim of providing a number of flexible courses within the subjects of Artificial intelligence (AI), Design, and Innovation management, for professionals. Our aim is to describe how the project is setup to create new learning opportunities, including the development process and co-creation with industry, the core structure and the pedagogical design. Furthermore, we would like to discuss both challenges and opportunities that come with this kind of project, as well as reflecting on early stage outcomes. © 2023, ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering.

  • 181.
    Johansson, Jörgen
    et al.
    University of Gothenburg, Gothenburg, Sweden.
    Thomsen, Michel
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Åkesson, Maria
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Public value creation and robotic process automation: normative, descriptive and prescriptive issues in municipal administration2023Ingår i: Transforming Government: People, Process and Policy, ISSN 1750-6166, E-ISSN 1750-6174, Vol. 17, nr 2, s. 177-191Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    Abstract Purpose – This paper aims to highlight problems and opportunities for introducing digital automation in public administration (PA) and to propose implications for public value creation of robotic process automation (RPA) through the perspective of good bureaucracy as a guiding framework.

    Design/methodology/approach – This conceptual paper addresses the purpose by applying three normative ideal types: Weber’s ideal type for a bureaucracy, new public management and public value management. This paper synthesizes an analytical framework in conducting case studies of the implementation of RPA systems in municipal administration.

    Findings – This paper contributes to new insights into public value creation and digital automation. The following four implications are proposed: the deployment of RPA in municipal administration should emphasize that organizing administrative tasks is essentially a political issue; include considerations based on a wellgrounded analysis in which policy areas that are suitable for RPA; to pay attention to issues on legal certainty, personal integrity, transparency and opportunities to influence automated decisions; and that the introduction of RPA indicates a need to develop resources concerning learning and knowledge in the municipal administration.

    Originality/value – This paper is innovative, as it relates normative, descriptive and prescriptive issues on the developing of digital automation in PA. The conceptual approach is unusual in studies of digitalization in public activities. © 2022, Jörgen Johansson, Michel Thomsen and Maria Åkesson.

    Ladda ner fulltext (pdf)
    fulltext
  • 182.
    Akbari, Maryam
    et al.
    Iran University of Science and Technology, Tehran, Iran.
    Mirzakuchaki, Sattar
    Iran University of Science and Technology, Tehran, Iran.
    Fazeli, Mahdi
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Tarihi, Mohammad Reza
    Niroo Research Institute, Tehran, Iran.
    Pure Magnetic Memory-Based PUFs: A Secure and Lightweight Solution for IoT Devices2023Ingår i: Iranian Journal of Electrical and Electronic Engineering, ISSN 2383-3890, Vol. 19, nr 4, artikel-id 2944Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    In light of the growing prevalence of Internet of Things (IoT) devices, it has become essential to incorporate cryptographic protection techniques for high-security applications. Since IoT devices are resource-constraints in terms of power and area, finding cost-effective ways to enhance their security is necessary. Physical unclonable function (PUF) is considered a trusted hardware security mechanism that generates true and intrinsic randomness by extracting the inherent process variations of circuits. In this paper, a novel pure magnetic memory-based PUF is presented. The fundamental building blocks of the proposed PUF design are magnetic devices, the so-called mCells. These magnetoresistive devices exclusively utilize Magnetic Tunnel Junction (MTJ) components. Using purely MTJ in the main memory and sense amplifier in the proposed PUF leads to high randomness, high reliability, low power, and ultra-compact occupation area. The Monte Carlo HSPICE simulation results demonstrate that the proposed PUF achieves a uniqueness of 49.89%, uniformity of 50.02 %, power consumption of 1.43 µW, and an area occupation of 0.01 µm2 per bit. © 2023, Iran University of Science and Technology. All rights reserved.

  • 183.
    Qu, Zhiguo
    et al.
    Nanjing University of Information Science and Technology, Nanjing, China; Nanjing University of Information Science and Technology, Nanjing, China.
    Li, Yang
    Nanjing University of Information Science and Technology, Nanjing, China; Nanjing University of Information Science and Technology, Nanjing, China.
    Tiwari, Prayag
    Högskolan i Halmstad, Akademin för informationsteknologi.
    QNMF: A quantum neural network based multimodal fusion system for intelligent diagnosis2023Ingår i: Information Fusion, ISSN 1566-2535, E-ISSN 1872-6305, Vol. 100, artikel-id 101913Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    The Internet of Medical Things (IoMT) has emerged as a significant research area in the medical field, enabling the transmission of various types of data to the cloud for analysis and diagnosis. Fusing data from multiple modalities can enhance accuracy but requires substantial computing power. Theoretically, quantum computers can rapidly process large volumes of high-dimensional medical data. Despite accelerated developments in quantum computing, research on quantum machine learning (QML) for multimodal data processing remains limited. Considering these factors, this paper presents a quantum neural network-based multimodal fusion system for intelligent diagnosis (QNMF) that can process multimodal medical data transmitted by IoMT devices, fuse data from different modalities, and improve the performance of intelligent diagnosis. This system employs a quantum convolutional neural network (QCNN) to efficiently extract features from medical images. These QCNN-based features are then fused with other modality features (such as blood test results or breast cell slices), and used to train an effective variational quantum classifier (VQC) for intelligent diagnosis. The experimental results demonstrate that a QCNN can effectively extract image data features. Furthermore, QNMF achieved an accuracy of 97.07% and 97.61% on breast cancer diagnosis and Covid-19 diagnosis experiments, respectively. In addition, the QNMF exhibits strong quantum noise robustness. © 2023 Elsevier B.V.

  • 184.
    Qu, Zhiguo
    et al.
    Nanjing University Of Information Science And Technology, Nanjing, China.
    Shi, Wenke
    Nanjing University Of Information Science And Technology, Nanjing, China.
    Tiwari, Prayag
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Quantum conditional generative adversarial network based on patch method for abnormal electrocardiogram generation2023Ingår i: Computers in Biology and Medicine, ISSN 0010-4825, E-ISSN 1879-0534, Vol. 166, s. 1-13, artikel-id 107549Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    To address the scarcity and class imbalance of abnormal electrocardiogram (ECG) databases, which are crucial in AI-driven diagnostic tools for potential cardiovascular disease detection, this study proposes a novel quantum conditional generative adversarial algorithm (QCGAN-ECG) for generating abnormal ECG signals. The QCGAN-ECG constructs a quantum generator based on patch method. In this method, each sub-generator generates distinct features of abnormal heartbeats in different segments. This patch-based generative algorithm conserves quantum resources and makes QCGAN-ECG practical for near-term quantum devices. Additionally, QCGAN-ECG introduces quantum registers as control conditions. It encodes information about the types and probability distributions of abnormal heartbeats into quantum registers, rendering the entire generative process controllable. Simulation experiments on Pennylane demonstrated that the QCGAN-ECG could generate completely abnormal heartbeats with an average accuracy of 88.8%. Moreover, the QCGAN-ECG can accurately fit the probability distribution of various abnormal ECG data. In the anti-noise experiments, the QCGAN-ECG showcased outstanding robustness across various levels of quantum noise interference. These results demonstrate the effectiveness and potential applicability of the QCGAN-ECG for generating abnormal ECG signals, which will further promote the development of AI-driven cardiac disease diagnosis systems. The source code is available at github.com/VanSWK/QCGAN_ECG. © 2023 The Author(s)

  • 185.
    Qu, Zhiguo
    et al.
    Nanjing University Of Information Science And Technology, Nanjing, China.
    Zhang, Zhexi
    Nanjing University Of Information Science And Technology, Nanjing, China.
    Liu, Bo
    Hubei University Of Science And Technology, Xianning, China.
    Tiwari, Prayag
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Ning, Xin
    Chinese Academy Of Sciences, Beijing, China.
    Muhammad, Khan
    Sungkyunkwan University, Seoul, South Korea.
    Quantum detectable Byzantine agreement for distributed data trust management in blockchain2023Ingår i: Information Sciences, ISSN 0020-0255, E-ISSN 1872-6291, Vol. 637, artikel-id 118909Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    No system entity within a contemporary distributed cyber system can be entirely trusted. Hence, the classic centralized trust management method cannot be directly applied to it. Blockchain technology is essential to achieving decentralized trust management, its consensus mechanism is useful in addressing large-scale data sharing and data consensus challenges. Herein, an n-party quantum detectable Byzantine agreement (DBA) based on the GHZ state to realize the data consensus in a quantum blockchain is proposed, considering the threat posed by the growth of quantum information technology on the traditional blockchain. Relying on the nonlocality of the GHZ state, the proposed protocol detects the honesty of nodes by allocating the entanglement resources between different nodes. The GHZ state is notably simpler to prepare than other multi-particle entangled states, thus reducing preparation consumption and increasing practicality. When the number of network nodes increases, the proposed protocol provides better scalability and stronger practicability than the current quantum DBA. In addition, the proposed protocol has the optimal fault-tolerant found and does not rely on any other presumptions. A consensus can be reached even when there are n−2 traitors. The performance analysis confirms viability and effectiveness through exemplification. The security analysis also demonstrates that the quantum DBA protocol is unconditionally secure, effectively ensuring the security of data and realizing data consistency in the quantum blockchain. © 2023 The Authors

  • 186.
    Lou, Chuyue
    et al.
    School of Automation, Wuhan University of Technology, Wuhan, China.
    Atoui, M. Amine
    Högskolan i Halmstad, Akademin för informationsteknologi, Centrum för forskning om tillämpade intelligenta system (CAISR).
    Li, Xiangshun
    School of Automation, Wuhan University of Technology, Wuhan, China.
    Recent deep learning models for diagnosis and health monitoring: a review of researches and future challenges2023Ingår i: Transactions of the Institute of Measurement and Control, ISSN 0142-3312, E-ISSN 1477-0369Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    As an important branch of machine learning, deep learning (DL) models with multiple hidden layer structures have the ability to extract highly representative features from the input. At present, fault detection and diagnosis (FDD) and health monitoring solutions developed based on DL models have received extensive attention in academia and industry along with the rapid improvement of computing power. Therefore, this paper focuses on a comprehensive review of DL model–based FDD and health monitoring schemes in view of common problems of industrial systems. First, brief theoretical backgrounds of basic DL models are introduced. Then, related publications are discussed about the development of DL and graphical models in the industrial context. Afterwards, public data sets are summarized, which are associated with several research papers. More importantly, suggestions on DL model–based diagnosis and health monitoring solutions and future developments are given. Our work will have a positive impact on the selection and design of FDD solutions based on DL and graphical models in the future. © The Author(s) 2023.

    Ladda ner fulltext (pdf)
    fulltext
  • 187.
    Ebbesson, Esbjörn
    et al.
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Fors, Vaike
    Högskolan i Halmstad, Akademin för informationsteknologi, Halmstad Embedded and Intelligent Systems Research (EIS).
    Retaining ways of co-creation2023Ingår i: ECIS 2023 Research Papers: ECIS 2023, European Conference of Information Systems, Kristianstand, Norway, 2023Konferensbidrag (Refereegranskat)
    Abstract [en]

    The design space of future mobility services is considered a wicked problem, as many stakeholders from the public and private sectors need to collaborate to create sustainable future services. Recent years have shown a growing interest in utilizing urban living labs (ULL) and similar quadruple helix approaches toward addressing wicked design challenges. However, when engaging in co-creation through living labs, many actors also see potential in adapting methodology and new ways-of-doing, to appropriate it and improve readiness for tackling other wicked challenges. The article draws upon a ULL initiative in the mobility service context to explore the main challenges for ULL partners to retain the ways-of-doing that develops in co-creation activities. Through our study, we identified that cocreation needs to be grounded in the known, to facilitate search and co-appropriation of the unknown as key for retaining ways-of-doing in ULL initiatives.

    Ladda ner fulltext (pdf)
    Full text article
  • 188.
    Aslam, Muhammad Shamrooz
    et al.
    China University of Mining And Technology, Xuzhou, China; Guangxi University of Science and Technology, Liuzhou, China.
    Tiwari, Prayag
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Pandey, Hari Mohan
    Bournemouth University, Bournemouth, United Kingdom.
    Band, Shahab S.
    National Yunlin University of Science and Technology, Douliou, Taiwan.
    Robust stability analysis for class of Takagi-Sugeno (T-S) fuzzy with stochastic process for sustainable hypersonic vehicles2023Ingår i: Information Sciences, ISSN 0020-0255, E-ISSN 1872-6291, Vol. 641, artikel-id 119044Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    Recently, the rapid development of Unmanned Aerial Vehicles (UAVs) enables ecological conservation, such as low-carbon and “green” transport, which helps environmental sustainability. In order to address control issues in a given region, UAV charging infrastructure is urgently needed. To better achieve this task, an investigation into the T–S fuzzy modeling for Sustainable Hypersonic Vehicles (SHVs) with Markovian jump parameters and H∞ attitude control in three channels was conducted. Initially, the reentry dynamics were transformed into a control–oriented affine nonlinear model. Then, the original T–S local modeling method for SHV was projected by primarily referring to Taylor's expansion and fuzzy linearization methodologies. After the estimation of precision and controller complexity was assumed, the fuzzy model for jump nonlinear systems mainly consisted of two levels: a crisp level and a fuzzy level. The former illustrates the jumps, and the latter a fuzzy level that represents the nonlinearities of the system. Then, a systematic method built in a new coupled Lyapunov function for a stochastic fuzzy controller was used to guarantee the closed–loop system for H∞ gain in the presence of a predefined performance index. Ultimately, numerical simulations were conducted to show how the suggested controller can be successfully applied and functioned in controlling the original attitude dynamics. © 2023 Elsevier Inc.

  • 189.
    Kochenborger Duarte, Eduardo
    et al.
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Erneberg, Mikael
    H&E Solutions AB, Stockholm, Sweden.
    Pignaton de Freitas, Edison
    Federal University of Rio Grande do Sul Porto Alegre, Brazil.
    Bellalta, Boris
    Universitat Pompeu Fabra, Barcelona, Spain.
    Vinel, Alexey
    Karlsruhe Institute of Technology, Karlsruhe, Germany.
    SafeSmart 6G: The Future of Emergency Vehicle Traffic Light Preemption2023Ingår i: 2023 2nd International Conference on 6G Networking (6GNet), IEEE, 2023Konferensbidrag (Refereegranskat)
    Abstract [en]

    This paper delves into the utilization of Vehicular Ad-hoc Networks (VANETs) in emergency vehicle warning systems in the era of 6G. The proposed system, named SafeSmart 6G, will leverage VANET-based vehicle-to-infrastructure Communication powered by 6G to exchange data between traffic lights and emergency vehicles, enhancing safety and reducing response times. SafeSmart 6G will predict the arrival time of emergency vehicles at intersections using historical data and AI-driven analytics, requesting signal preemption for the chosen route. The paper discusses the potential benefits and challenges that might arise from the use of 6G in emergency scenarios. © 2023 IEEE.

  • 190.
    Kochenborger Duarte, Eduardo
    et al.
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Erneberg, Mikael
    H&E Solutions Ab, Stockholm, Sweden.
    Freitas, Edison Pignaton De
    Universidade Federal Do Rio Grande Do Sul, Porto Alegre, Brazil.
    Bellalta, Boris
    Universitat Pompeu Fabra, Barcelona, Spain.
    Vinel, Alexey
    Karlsruhe Institute Of Technology, Karlsruhe, Germany.
    SafeSmart: A VANET-LTE-based solution for faster and safer response in critical situations2023Ingår i: IEEE Conference on Standards for Communications and Networking: 2023, Piscataway, NJ: IEEE, 2023, s. 47-53Konferensbidrag (Refereegranskat)
    Abstract [en]

    This paper discusses the use of Vehicular Adhoc Networks (VANETs) for traffic light preemption in emergency scenarios. The proposed system, called SafeSmart, utilizes VANET-based vehicle-to-infrastructure communication to exchange data between traffic lights and emergency vehicles, improving safety and saving time. SafeSmart attempts to predict the arrival time of emergency vehicles at intersections using historical data and requests signal preemption for the selected route. This paper describes and evaluates the proposed approach through simulations using state-of-the-art simulators SUMO and OMNeT++ and real-world traffic data (Luxembourg SUMO Traffic (LuST) Scenario). The results demonstrate improved trip times and increased safety for emergency vehicles and general public on the road. © 2023 IEEE.

  • 191.
    Al Khatib, Sultan M.
    et al.
    Al-balqa Applied University, Al Salt, Jordan.
    Alkharabsheh, Khalid
    Al-balqa Applied University, Al Salt, Jordan.
    Alawadi, Sadi
    Högskolan i Halmstad, Akademin för informationsteknologi. Uppsala University, Uppsala, Sweden.
    Selection of human evaluators for design smell detection using dragonfly optimization algorithm: An empirical study2023Ingår i: Information and Software Technology, ISSN 0950-5849, E-ISSN 1873-6025, Vol. 155, artikel-id 107120Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    Context: Design smell detection is considered an efficient activity that decreases maintainability expenses and improves software quality. Human context plays an essential role in this domain. Objective: In this paper, we propose a search-based approach to optimize the selection of human evaluators for design smell detection. Method: For this purpose, Dragonfly Algorithm (DA) is employed to identify the optimal or near-optimal human evaluator's profiles. An online survey is designed and asks the evaluators to evaluate a sample of classes for the presence of god class design smell. The Kappa-Fleiss test has been used to validate the proposed approach. Results: The results show that the dragonfly optimization algorithm can be utilized effectively to decrease the efforts (time, cost ) of design smell detection concerning the identification of the number and the optimal or near-optimal profile of human experts required for the evaluation process. Conclusions: A Search-based approach can be effectively used for improving a god-class design smell detection. Consequently, this leads to minimizing the maintenance cost. © 2022 The Author(s)

  • 192.
    Liang, Guojun
    et al.
    Faculty of Innovation Engineering, Macau University of Science and Technology, Macau, China.
    U, Kintak
    Faculty of Innovation Engineering, Macau University of Science and Technology, Macau, China.
    Ning, Xin
    Laboratory of Artificial Neural Networks and High Speed Circuits, Institute of Semiconductors, Chinese Academy of Sciences, China.
    Tiwari, Prayag
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Nowaczyk, Sławomir
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Kumar, Neeraj
    School of Computer Science, University of Petroleum and Energy Studies, Dehradun, Uttarakhand, India; Lebanese American University, Beirut, Lebanon; King Abdulaziz University, Jeddah, Saudi Arabia.
    Semantics-aware Dynamic Graph Convolutional Network for Traffic Flow Forecasting2023Ingår i: IEEE Transactions on Vehicular Technology, ISSN 0018-9545, E-ISSN 1939-9359, Vol. 72, nr 6, s. 7796-7809Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    Traffic flow forecasting is a challenging task due to its spatio-temporal nature and the stochastic features underlying complex traffic situations. Currently, Graph Convolutional Network (GCN) methods are among the most successful and promising approaches. However, most GCNs methods rely on a static graph structure, which is generally unable to extract the dynamic spatio-temporal relationships of traffic data and to interpret trip patterns or motivation behind traffic flows. In this paper, we propose a novel Semantics-aware Dynamic Graph Convolutional Network (SDGCN) for traffic flow forecasting. A sparse, state-sharing, hidden Markov model is applied to capture the patterns of traffic flows from sparse trajectory data; this way, latent states, as well as transition matrices that govern the observed trajectory, can be learned. Consequently, we can build dynamic Laplacian matrices adaptively by jointly considering the trip pattern and motivation of traffic flows. Moreover, high-order Laplacian matrices can be obtained by a newly designed forward algorithm of low time complexity. GCN is then employed to exploit spatial features, and Gated Recurrent Unit (GRU) is applied to exploit temporal features. We conduct extensive experiments on three real-world traffic datasets. Experimental results demonstrate that the prediction accuracy of SDGCN outperforms existing traffic flow forecasting methods. In addition, it provides better explanations of the generative Laplace matrices, making it suitable for traffic flow forecasting in large cities and providing insight into the causes of various phenomena such as traffic congestion. The code is publicly available at https://github.com/gorgen2020/SDGCN. © 2023 IEEE.

    Ladda ner fulltext (pdf)
    fulltext
  • 193.
    Delooz, Quentin
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Sensor Data Sharing in V2X Communications: Protocol Design and Performance Optimization of Collective Perception2023Doktorsavhandling, sammanläggning (Övrigt vetenskapligt)
    Abstract [en]

    Sensor data sharing involves exchanging sensor data among multiple devices, systems, or platforms through various means, such as wired or wireless communication, cloud storage, and distributed computing. In Vehicle-to-Everything (V2X) communication, sensor data sharing is known as Collective Perception (CP). V2X Collective Perception is the principle of exchanging sensor data among V2X-capable stations, such as vehicles, vulnerable road users, or roadside units, by exchanging lists of perceived objects in the allocated 5.9 GHz frequency band for road safety and traffic efficiency. An object can be anything relevant to traffic safety and is described using its characteristics such as position, heading, and velocity. Objects are detected thanks to sensors such as cameras, LiDARs, and radars mounted on V2X stations. This thesis investigates the message generation rule for CP, specifically how often and with which objects a Collective Perception Message (CPM) should be generated for transmission. The contained studies focus on the challenges posed by the limited bandwidth available in the 5.9 GHz channel against the object selection for inclusion in CPMs. In the first part of the realized studies, the protocol design and the requirements of CP are comprehended from the network and application-related aspects, concluding that the process of filtering objects is necessary to control the channel usage of CP. Moreover, results show that object filtering is only beneficial in high-traffic density scenarios and should not be applied when channel resources are plenty available. In the second part, methods are developed and assessed to adapt the object filtering mechanism to the available channel resources and control information redundancy, i.e., controlling the number of vehicles transmitting updates about the same objects. Through a combination of theoretical analysis, large-scale simulations, and experimental evaluation, this thesis provides a better understanding of the requirements of CP for object filtering and shows the benefits of a developed novel algorithm to adapt object filtering to the available channel resources. Additionally, it elaborates on new metrics and provides a requirements analysis and performance assessment of selected information redundancy reduction techniques. Finally, the results show that combining both approaches enables efficient control of information redundancy while allowing efficient channel resource usage.

    Ladda ner fulltext (pdf)
    fulltext
  • 194.
    Koutsikouri, Dina
    et al.
    Gothenburg University, Gothenburg, Sweden.
    Hylving, Lena
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Lindberg, Susanne
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Bornemark, Jonna
    Södertörn University, Huddinge, Sweden.
    Seven Elements of Phronesis: A Framework for Understanding Judgment in Relation to Automated Decision-Making2023Ingår i: Proceedings of the 56th Hawaii International Conference on System Sciences / [ed] Tung X. Bui, IEEE Computer Society, 2023, Vol. 56, s. 5292-5301Konferensbidrag (Refereegranskat)
    Abstract [en]

    This conceptual paper aims to explore judgment in the context of automated decision-making systems (ADS). To achieve this, we adopt a modern version of Aristotle’s notion of phronesis to understand judgment. We delineate seven elements of judgment which provide insights into what humans are better at, and what AI is better at in relation to automated decision-making. These elements are sources of knowledge that guide action including not-knowing, emotions, sensory perception, experience, intuition, episteme, and techne. Our analysis suggests that most of these attributes are not transferable to AI systems, because judgment in human decision-making requires the integration of all which involves considering the contextual and affective resources of phronesis, and the competence to make value judgments. The paper contributes to unpack human judgment capacities and what needs to be cultivated to achieve ‘good’ AI systems that serves humanity as well as guiding future information systems researchers to explore human-AI judgment further.

  • 195.
    Delooz, Quentin
    et al.
    Högskolan i Halmstad, Akademin för informationsteknologi. Technische Hochschule Ingolstadt, CARISSMA, Ingolstadt, Germany.
    Festag, Andreas
    Technische Hochschule Ingolstadt, CARISSMA, Ingolstadt, Germany; Fraunhofer IVI, Technische Universität Dresden, Dresden, Germany.
    Vinel, Alexey
    Högskolan i Halmstad, Akademin för informationsteknologi. Karlsruhe Institute of Technology, Universität Karlsruhe, Karlsruhe, Germany.
    Lobo, Silas C.
    Technische Hochschule Ingolstadt, CARISSMA, Ingolstadt, Germany.
    Simulation-based Performance Optimization of V2X Collective Perception by Adaptive Object Filtering2023Ingår i: 2023 IEEE Intelligent Vehicles Symposium (IV), Piscataway, NJ: IEEE, 2023Konferensbidrag (Refereegranskat)
    Abstract [en]

    V2X Collective Perception is the principle of exchanging sensor data among V2X-capable stations, such as vehicles or roadside units, by exchanging lists of perceived objects in the 5.9 GHz frequency band for road safety and traffic efficiency. An object can be anything relevant to traffic safety, e.g.,vehicles or pedestrians. The current standardization of Collective Perception in Europe considers filtering objects for transmission based on their locally perceived dynamics and freshness to preserve channel resources. However, two remaining problems of object filtering are: information redundancy and adapting object filtering to the available channel resources. In this paper, we combine redundancy mitigation and congestion control-aware filtering. We evaluate the performance of the resulting object filtering techniques by realizing realistic, large-scale simulations of a mid-size city in Germany. We assess the performance using ascoring metric. The results show better information redundancy control and adjustable channel usage for object filtering. © Copyright 2023 IEEE

  • 196.
    Gong, Zijun
    et al.
    University of Waterloo, Waterloo, Canada; The Hong Kong University of Science and Technology (HKUST), Guangzhou, China; Department of ECE, HKUST, Hong Kong, China.
    Jiang, Fan
    Högskolan i Halmstad, Akademin för informationsteknologi. Laboratory (PCL), Shenzhen, China.
    Li, Cheng
    Simon Fraser University, Burnaby, Canada.
    Shen, Xuemin
    University of Waterloo, Waterloo, Canada.
    Simultaneous Localization and Communications With Massive MIMO-OTFS2023Ingår i: IEEE Journal on Selected Areas in Communications, ISSN 0733-8716, E-ISSN 1558-0008, Vol. 41, nr 12, s. 3908-3924Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    Next generation cellular network is expected to provide the simultaneous high-accuracy localization and ultra-reliable communication services, even in high mobility scenarios. To that end, the novel orthogonal time frequency space (OTFS) modulation has been developed as a promising physical-layer transmission technique, evident by the outstanding performance in terms of robustness against time-frequency selective fading over the orthogonal frequency division multiplexing (OFDM) counterpart. However, when OTFS meets massive multiple-input multiple-output (MIMO), the specific conditions, under which the delay-Doppler (DD) domain channel model holds, are not identified. In addition, the channel estimation and localization performance in such system is rarely studied. In this work, we target at these new challenges, and conduct comprehensive modelling, performance analysis, and algorithm design for massive MIMO-OTFS based simultaneous localization and communications. Specifically, we derive new channel models for the massive MIMO-OTFS system, which captures both time-frequency dispersion and spatial wideband effects. The specific conditions, under which the new models hold has been unveiled as well. Based on the new models, we establish the theoretical foundations for channel estimation and localization, by deriving the Cramer-Rao lower bounds of channel parameter and location estimation errors. Such bounds have been achieved with the newly designed low-complexity channel estimation and localization algorithms. Numerical simulations of the proposed framework with prevailing pulse functions are also conducted and the results validate the proposed designs and analysis. © 1983-2012 IEEE.

  • 197.
    Amjad, Iqbal
    et al.
    Högskolan i Halmstad, Akademin för informationsteknologi. Institut National de la Recherche Scientifique, Montreal, Canada.
    Al-Hasan, Muath
    Al Ain University, Al Ain, United Arab Emirates.
    Mabrouk, Ismail Ben
    Durham University, Durham, United Kingdom.
    Denidni, Tayeb A.
    Institut National de la Recherche Scientifique, Montreal, Canada.
    Simultaneous Transmit and Receive Self-Duplexing Antenna for Head Implants2023Ingår i: IEEE Transactions on Antennas and Propagation, ISSN 0018-926X, E-ISSN 1558-2221, Vol. 71, nr 11, s. 8592-8601Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    A novel dual-port, dual-band self-duplexing implantable antenna for head implants is presented in this article. It operates at 915 MHz Industrial, Scientific, and Medical (ISM) band (when Port 1 is active) and 1420 MHz Wireless Medical Telemetry Service (WMTS) band (when Port2 is active). The proposed antenna is kept inside a flat device and simulated in a human head model. The miniaturization of the proposed antenna is achieved using a high dielectric substrate, shorted pins, and multiple capacitive slots. Consequently, it occupies a compact volume of 7.13 x 8.9 x 0.13 = 8.24 mm(3). The coupling between both radiators is enhanced by printing it on a thin substrate and placing vias between them. As a result, an isolation level better than 31.4 dB is achieved. It has peak realized gains of -18.94 and -17.06 dBi at 915 and 1420 MHz, respectively. The link budget analysis and specific absorption rate (SAR) are performed, showing promising results. The proposed concept is practically validated by measuring its performance inside the minced pork meat. Furthermore, the simultaneous transmit and receive concept is practically verified with the aid of software-defined radio (SDR). The main advantages of this antenna are its compact size, low coupling level, independently controllable bands, and simultaneous transmission and reception of signals without using a multiplexer circuit. © 2023 IEEE.

  • 198.
    Amininasab, Mehdi
    et al.
    Independent Researcher, Tehran, Iran.
    Patooghy, Ahmad
    North Carolina A&T State University, North Carolina, United States.
    Fazeli, Mahdi
    Högskolan i Halmstad, Akademin för informationsteknologi.
    SingAll: Scalable Control Flow Checking for Multi-Process Embedded Systems2023Ingår i: 2023 13th International Conference on Computer and Knowledge Engineering (ICCKE), IEEE, 2023, s. 42-47Konferensbidrag (Refereegranskat)
    Abstract [en]

    Reliability concerns of embedded systems are traditionally resolved by software-based control flow checking (CFC) methods where the execution flow of the processor is monitored to detect and compensate flow violations. Traditional CFC methods may lose their efficiency when it comes to multiprocessing embedded systems. In this paper, we introduce and validate a novel flow error model for multiprocessing embedded systems. Further, we propose a holistic CFC system which performs the flow checking of the processes of interest. The proposed CFC checking introduces the concept of a single monitoring process intended to check the execution flow of as many processes as wanted within an multiprocessing embedded system. Proposed solution does not introduce any substantial overheads in performance and memory consumption. Even more important is method's insensitivity to the number of checked processes. Our wide evaluations show the average performance overhead of 13.77%, average code-size overhead of 51.71%, and the average memory overhead of 1.95% on the Mibench benchmark suite. Results of fault injections confirm that the proposed CFC method successfully detects more than 95% of flow errors including our newly defined error model. © 2023 IEEE.

  • 199.
    Khan, Hameed Ullah
    et al.
    Department of Computer Science, COMSATS University Islamabad, Islamabad, Pakistan.
    Raza, Basit
    Department of Computer Science, COMSATS University Islamabad, Islamabad, Pakistan.
    Shah, Munawar Hussain
    Pathology Department, Nishtar Medical University, Multan, Pakistan.
    Usama, Syed Muhammad
    Post Graduate Resident Surgeon at College of Physicians and Surgeons Pakistan (CPSP), Karachi, Pakistan.
    Tiwari, Prayag
    Högskolan i Halmstad, Akademin för informationsteknologi.
    Band, Shahab S.
    Future Technology Research Center, College of Future, National Yunlin University of Science and Technology, Yunlin, Douliou, Taiwan.
    SMDetector: Small mitotic detector in histopathology images using faster R-CNN with dilated convolutions in backbone model2023Ingår i: Biomedical Signal Processing and Control, ISSN 1746-8094, E-ISSN 1746-8108, Vol. 81, artikel-id 104414Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    Breast cancer is one of the most common cancer types among women, and it is a deadly disease caused by the uncontrolled proliferation of cells. Pathologists face a challenging issue of mitotic cell identification and counting during manual detection and identification of cancer. Artificial intelligence can help the medical professional with early, quick, and accurate diagnosis of breast cancer. Consequently, the survival rate will be improved and mortality rate will be decreased. Different deep learning techniques are used in computational pathology for cancer diagnosis. In this study, the SMDetector is proposed which is a deep learning model for detecting small objects such as mitotic and non-mitotic nuclei. This model employs dilated layers in the backbone to prevent small objects from disappearing in the deep layers. The purpose of the dilated layers in this model is to reduce the size gap between the image and the objects it contains. Region proposal network is optimized to accurately identify small objects. The proposed model yielded overall average precision (AP) of 50.31% and average recall (AR) of 55.90% that outperforms the existing standard object detection models on ICPR 2014 (Mitos-Atypia-14) dataset. To best of our knowledge the proposed model is state-of-the-art model for precision and recall of mitotic object detection on ICPR 2014 (Mitos-Atypia-14) dataset. The proposed model has achieved average precision for mitotic nuclei 68.49%, average recall for mitotic nuclei 59.86% and f-measure 63.88%. © 2022 The Authors

  • 200.
    Jeddi Abdarloo, Hossein
    et al.
    Högskolan i Halmstad, Akademin för informationsteknologi. Solid State Physics and NanoLund, Lund University, Lund, Sweden.
    Witzigmann, Bernd
    Institute for Optoelectronics, Friedrich-Alexander Universität Erlangen-Nürnberg, Erlangen, Germany.
    Adham, Kristi
    Solid State Physics and NanoLund, Lund University, Lund, Sweden.
    Hrachowina, Lukas
    Solid State Physics and NanoLund, Lund University, Lund, Sweden.
    Borgström, Magnus T.
    Solid State Physics and NanoLund, Lund University, Lund, Sweden.
    Pettersson, Håkan
    Högskolan i Halmstad, Akademin för informationsteknologi. Solid State Physics and NanoLund, Lund University, Lund, Sweden.
    Spectrally Tunable Broadband Gate-All-Around InAsP/InP Quantum Discs-in-Nanowire Array Phototransistors with a High Gain-Bandwidth Product2023Ingår i: ACS Photonics, E-ISSN 2330-4022, Vol. 10, nr 6, s. 1748-1755Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    High-performance broadband photodetectors offering spectral tunability and a high gain-bandwidth product are crucial in many applications. Here, we report on a detailed experimental and theoretical study of three-terminal phototransistors comprised of three million InP nanowires with 20 embedded InAsP quantum discs in each nanowire. A global, transparent ITO gate all around the nanowires facilitates a radial control of the carrier concentration by more than two orders of magnitude. The transfer characteristics reveal two different transport regimes. In the subthreshold region, the photodetector operates in a diffusion mode with a distinct onset at the bandgap of InP. At larger gate biases, the phototransistor switches to a drift mode with a strong contribution from the InAsP quantum discs. Besides an unexpected spectral tunability, the detector exhibits a state-of-the-art responsivity, reaching around 100 A/W (638 nm/20 μW) @ VGS = 1.0 V/VDS = 0.5 V with a gain-bandwidth product of around 1 MHz, in excellent agreement with a comprehensive real-device model. © 2023 The Authors. Published by American Chemical Society.

1234567 151 - 200 av 2367
RefereraExporteraLänk till träfflistan
Permanent länk
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annat format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annat språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf