This paper proposes a near-field Euler rotation-based technique to transform the far-field calculation of arbitrary conformal arrays into the pattern-multiplication form. Additionally, the calculation is greatly accelerated by a three-dimensional virtual equivalent source expansion (3-D VESE) technique and a layer-wise two-dimensional fast Fourier transform (2-D FFT) technique. Computational complexity analysis and several numerical examples validate the advantages of the proposed solution in terms of the efficiency and accuracy.

Our study delves into the challenges of emotion recognition through electroencephalogram (EEG) signals in brain-computer interface systems. Recognizing the limitations of existing methods in accurately capturing intricate emotional patterns in EEG data, we propose a novel approach using asymmetric windowing recurrence plots (AWRP). This technique was designed to enhance the efficiency and accuracy of emotion recognition by encoding EEG signals into detailed image representations that are suitable for advanced deep neural network analysis. Through empirical validations using benchmark datasets (DEAP and SEED), our method demonstrated significant improvements in classification accuracies, notably outperforming existing state-of-the-art methodologies. These findings not only contribute to the field of EEG-based emotion recognition, but also present a novel perspective that can guide future research in neural system analysis and rehabilitation engineering.Â

Construction and analysis of functional brain networks (FBNs) with rs-fMRI is a promising method to diagnose functional brain diseases. Nevertheless, the existing methods suffer from several limitations. First, the functional connectivities (FCs) of the FBN are usually measured by the temporal co-activation level between rs-fMRI time series from regions of interest (ROIs). While enjoying simplicity, the existing approach implicitly assumes simultaneous co-activation of all the ROIs, and models only their synchronous dependencies. However, the functional coactivation is not necessarily always synchronous due to the time lag of information flow and cross-time interactions between ROIs. Therefore, it is desirable to model the asynchronous functional interactions. Second, the traditional methods usually construct FBNs at individual level for feature extraction and classification, leading to large variability and degraded diagnosis accuracy when modeling asynchronous FBN. Third, the FBN construction and analysis are conducted in two independent steps without joint alignment for the target diagnosis task. To address the first limitation, this paper proposes an effective sliding-window-based method to model spatiotemporal FCs in Transformer. Regarding the second limitation, we propose to learn common and individual FBNs adaptively with the common FBN as prior knowledge, thus alleviating the variability and enabling the network to focus on the individual disease-specific asynchronous FCs. To address the third limitation, the common and individual asynchronous FBNs are built and analyzed by an integrated network, enabling end-to-end training and improving the flexibility and discriminativity. The effectiveness of the proposed method is consistently demonstrated on three data sets for MCI diagnosis.

As the Internet of Things (IoT) evolves rapidly across various industries, the number of IoT protocols and applications is growing with vast number of heterogeneous components and entities. In setups with thousands of IoT devices, manual deployment of applications and device registration become impractical due to their time-consuming and costly nature, as well as the requirement for background knowledge of IoT devices and protocols. Furthermore, IoT devices often have resource constraints that prevent them from running complex software. Therefore, there is a significant need to enhance and optimize edge computing systems for IoT, making them suitable and dynamic for automated IoT device registration and heterogeneous application deployment. In this paper, we present an edge-based framework designed to facilitate the automated registration of diverse wireless IoT devices and the deployment of IoT applications. To validate our approach, we use a smart irrigation system enhanced with a containerized machine learning model as a proof of concept. Our evaluation of the implemented prototype demonstrates that our system is scalable and feasible.

The rapid development of transportation electrification brings about the popularity of interconnected power-transportation systems (IPTS). However, the escalating frequency and uncertainty of natural hazards, such as typhoons, pose threats and potential damage to the operation of IPTS. Electric vehicles (EVs) can serve as mobile energy sources, whose proper scheduling in transportation networks can provide power support for the damaged power networks caused by natural hazards, thus enhancing the system’s resilience. This paper proposes a two-stage scenario-based scheduling framework using EVs for the restoration of an IPTS under natural hazard risks. In the first stage, EVs are pre-allocated and pre-charged at the charging stations to maximize their support potential against the predicted hazards; in the second stage, EVs are re-dispatched among different charging stations to help restore the power demands given the damaged IPTS topology. To address the real hazard scenarios and reduce the computational burden, a scenario generation approach indicating the real hazard’s impact on the IPTS is proposed followed by a scenario-reduction algorithm. Numerical experiments are conducted to validate the effectiveness of the proposed method based on the IEEE 33-bus distribution network and the Sioux Falls transportation network.

To solve the problem of path planning in unknown dynamic environment, this paper proposes BRRT*-DWA algorithm with Adaptive Monte Carlo Localization. Bidirectional Rapidly-exploring Random Tree Star(BRRT*) is used to generate an optimal global path plan, Dynamic Window Approach(DWA) is a local planner and Adaptive Monte Carlo Localization(AMCL) is used as a localization technique.

The growing trend of solitary living among the elderly and young, coupled with the high risk of falls leading to injuries and death, highlights the need for fall monitoring systems. Emphasizing individuals' privacy and comfort, these systems should rely on radar sensors instead of visual-based, acoustic-based, or wearable solutions. Current radar-based systems are yet to reach satisfactory real-world performance. This work proposes a radar-based fall detection system with superior performance in complex real-world scenarios while maintaining edge computing capabilities and minimum hardware resources. The proposed deep learning system achieved a recall of 98.99% and a precision of 99.32%. These unprecedented performance numbers are measured on the proposed dataset, which is the most real-life representative dataset in the literature. The system has 211.8k parameters and ~8.84 M Floating Point Operations (FLOPs), achieving an edge computing capability. Moreover, the efficient model construction eliminates redundant computation in real-time operation. Furthermore, this work proposes a novel metric that encompasses all dataset's quality aspects into a single number, which can be applied to all classification problems. This metric can then be used as a correction factor for performance metrics to put them in the context of the dataset used for testing.

Recently, there has been a growing interest in dielectric antennas and electromagnetic components based on periodic structures, since such components can be realized using new additive manufacturing technologies. The design of these components starts from the geometry of the unit cell, which determines not only the value of the effective refractive index but also the parameters of the entire considered component, like for example, the bandwidth of operation. For this reason, periodic structures with higher symmetry are of particular interest, since they can exhibit a wider bandwidth and other improved parameters. To analyze and design such components, we present a semi-analytical method for calculating the guiding properties of parallel-plate waveguides loaded with periodic glide-symmetric dielectric structures. It is shown that the presence of glide-symmetry can be simply included in the dispersion characteristic equation based on the transverse resonance condition.  The derived expressions reveal that due to glide-symmetry only the Bloch-Floquet modes of the same parity interact and contribute to the dispersion properties. The accuracy of the discussed method is verified through the analysis of several one-dimensional (1-D) and two-dimensional (2-D) structures suitable for the design of planar lens antennas with a large frequency band of operation.

In the realm of RPGs, creating immersive, persona-driven dialogues remains a challenge, especially in intricate settings like Call of Cthulhu (CoC). Existing methodologies often falter in portraying character personas within complex conversations accurately. To address this, we introduce a novel card-based framework, utilizing the advanced Baichuan-7B language model for tailored dialogue generation. Guided by detailed scene settings and character personas, Baichuan-7B exhibited a striking ability to craft context-aware dialogues for even unseen characters and scenarios. To assess the quality of these dialogues, we present an innovative metric, circumventing the traditional hurdles of human evaluations. Furthermore, insights into the attention mechanism shed light on the dynamics of information flow during dialogue creation. Collectively, our findings underscore the transformative potential of large language models in computational storytelling, particularly in RPG settings.

Federated learning (FL) is an emerging paradigm for decentralized training of machine learning models on distributed clients, without revealing the data to the central server. The learning scheme may be horizontal, vertical or hybrid (both vertical and horizontal). Most existing research work with deep neural network (DNN) modeling is focused on horizontal data distributions, while vertical and hybrid schemes are much less studied. In this paper, we propose a generalized algorithm FedEmb, for modeling vertical and hybrid DNN-based learning. The idea of our algorithm is characterized by higher inference accuracy, stronger privacy-preserving properties, and lower client-server communication bandwidth demands as compared with existing work. The experimental results show that FedEmb is an effective method to tackle both split feature & subject space decentralized problems. To be specific, there are 0.3% to 4.2% improvement on inference accuracy and 88.9 % time complexity reduction over baseline method.

Cardiovascular disease (CVD) poses a serious threat to individual health, highlighting the importance of early detection and proactive mitigation. With advancements in consumer electronics such as wearables and IoT, there exists an opportunity for enhanced CVD prediction for users. Machine Learning (ML) has been widely used to predict CVD risk (high/low) based on various factors and is a critical area of healthcare research. However, sharing data needed to predict CVD with machine learning models is challenging due to privacy concerns. Federated Learning (FL) enables distributed training of ML models without sharing raw data. However, it requires all training features to be available to all clients. To address this problem, we propose a Vertical Federated Learning (VFL) based method designed for use with consumer electronics platforms. The proposed method trains Neural Network (NN) model in a distributed manner where different data features are held by different parties. In this work, each party maintains a portion of separate data features, performs calculations on them locally, and then transfers only the necessary information to jointly train an NN model. We employ the proposed method for different use cases where the dataset features are distributed between: i) the patient and the hospital (2-splits); ii) the patient, the doctor, and the laboratory (3-splits); and iii) the patient, the doctor, the Electrocardiogram (ECG) center, and the laboratory (4-splits). Using a realistic dataset publicly available, we test the proposed methodology.

When the coding assignments step into more complicated and structured periods for individuals, how to cultivate the skills have become the urgent needs for people who are engaged in sophisticated coding development in both academics as well as industries, especially those who have some basic knowledge in their undergraduate studies. In this paper, a state-of-art interactive tool: Programming Decomposition Tool(PDT) is proposed to help people enhance their complicated coding skills and further to broaden their CS learning horizons. The tools have two unique features: 1. The learning interface is adaptive and dynamic with respect to the detection of the knowledge depth of users. 2. Difficult learning parts will be decomposed into several easier sessions for better acquisition.

Face recognition systems have made significant strides thanks to data-heavy deep learning models, but these models rely on large privacy-sensitive datasets. Recent work in facial analysis and recognition have thus started making use of synthetic datasets generated from GANs and diffusion based generative models. These models, however, lack fairness in terms of demographic representation and can introduce the same biases in the trained downstream tasks. This can have serious societal and security implications. To address this issue, we propose a methodology that generates unbiased data from a biased generative model using an evolutionary algorithm. We show results for StyleGAN2 model trained on the Flicker Faces High Quality dataset to generate data for singular and combinations of demographic attributes such as \textit{Black and Woman}. We generate a large racially balanced dataset of 13.5 million images, and show that it boosts the performance of facial recognition and analysis systems whilst reducing their biases.Â

As algorithms get better at their accuracy and computational efficiency, they invoke curiosity among the affected scientific communities to check if they can benefit from newer versions or not. Computer vision is one such domain that has observed rapid growth in terms of algorithmic advancements. The advent of deep learning was itself a catalyst for agile innovation. Coupled with rapid improvements in algorithms for object detection, the speed of innovation has become tremendous. And so there are many engineering and scientific disciplines that leverage object detection, any improvement in the latter has a ripple effect on the erstwhile. Computer Aided Laparoscopy (CAL) has come a long way due to object detection algorithms based on deep learning. Yet, every once in a while a new algorithm is released. It is tempting to see how a new algorithm may have affected the tool detection accuracy and efficiency in CAL. Recently version 8 of the famous You Look Only Once (YOLO) algorithm was released. Like all the past releases, it has been claimed that this version is better at detection accuracy as well as computational efficiency. This paper examines the performance of YOLOv8 at tool detection in a CAL context. We employed a well-known laparoscopy tool detection benchmark dataset in this research. Models with superior performance have been obtained as a result of this research. Models are superior in terms of both detection accuracy as well as inference speed. Moreover, the models are ready to be deployed into a production environment. The results that are reported in this paper are not only useful for the surgical community but also for the benchmarking of the YOLO algorithm.

本文揭示了 U-ViT 架构在扩散模型中尚未开发的潜力。该研究初步探索了 U-ViT 架构在多模态扩散模型的视觉生成任务中的贡献，并提出了一种专门为 U-ViT 架构设计的改进方案“FreeV”，标志着基于 U-Net 的 FreeU 增强框架在 Transformer 架构中的首次应用。FreeV 框架可显着提高生成质量，而无需额外的训练或微调。这项研究的关键见解在于平衡 U-ViT 中的主干网络、跳过连接和融合特征图的贡献，以充分利用两个组件的优势，同时规避 U-ViT 中特征融合的局限性。项目页面：https://github.com/GoldenFishes/FreeV

The development of our neoteric scientific fields has reached such a magnificent level that the benefits from innovation have extended to all regions, including those most distant. Modern information and communication network systems have greatly improved the overall effectiveness and performance of the wireless networks. With the emergence of cellular networks, the wireless communication network system has evolved from the first generation to the sixth generation (6G), which will provide an integrated framework for a variety of utilities and applications. The goal of 6G network systems is to improve channel capacity, low bit rate error, efficient signal transmission, and low latency while providing reliable connectivity and seamless communication. Various technologies are integrated with 6G network systems to provide controlled, efficient, and reliable communication. One emerging technology for 6G communication network systems is reconfigurable intelligent surfaces (RIS), which is ideal for smooth and controllable wireless signal transmission. RIS uses smart beamforming methods with the assistance of an electronic circuit controller to provide superior signal quality. This study presents a complete overview of RIS, including its hardware architecture, key characteristics, control mechanism, operating frequency, communication duplex mode, and operating mode. The study also elaborates on the RIS operational environment, advantages of implementing RIS, deployment, and different types of communication by utilizing RIS. Finally, we conclude with a discussion of challenges and future research directions to overcome limitations to achieve the goal of RIS implementation and RIS-aided communication for 6G communication systems.

Objective: In the context of pulmonary rehabilitation exercise training, wearable real-time monitoring of respiratory patterns may represent a valuable tool in increasing accessibility to treatment, as well as expanding the opportunities of treatment automation and locomotor-respiratory coupling. This work explores Hall effect sensing, paired with a permanent magnet, embedded in a chest strap. Methods: Experimental evaluation was performed considering as reference the gold-standard of respiratory monitoring, an airflow transducer, and performance was compared to another wearable device with analogous usability but with a different working principle - a piezoelectric sensor, also embedded in a chest strap. A total of 16 healthy participants performed 15 different activities, representative of pulmonary rehabilitation exercises, simultaneously using the three devices. Evaluation was performed based on detection of flow reversal events, as well as fiducials detection latency. Results: The proposed sensor shows comparable performance to the piezoelectric sensor with a mean ratio, precision, and recall of 1.10, 0.89, and 0.98, respectively, against 1.35, 0.71, and 0.96 of the piezoelectric sensor, overall also presenting consistently smaller latencies. Conclusion: The characterization of the proposed sensor also shows adequate monitoring capabilities for exercises that do not rely heavily on torso mobility, but may present a limitation when it comes to activities such as torso rotations and side stretches. Significance: This work expands the applicability of Hall effect sensors, demonstrating their use in the context of real-time respiratory monitoring.

Motion intent recognition for controlling prosthetic systems has long relied on machine learning algorithms. Artificial neural networks have shown great promise for solving such nonlinear classification tasks, making them a viable method for this purpose. To bring these advanced methods and algorithms beyond the confines of the laboratory and into the daily lives of prosthetic users, self-contained embedded systems are essential. However, embedded systems face constraints in size, computational power, memory footprint, and power consumption, as they must be non-intrusive and discreetly integrated into commercial prosthetic components. One promising approach to tackle these challenges is to use network quantization, which allows complying with limitations without significant loss in accuracy. Here, we compare network quantization performance for self-contained systems using TensorFlow Lite and the recently developed QKeras platform. Due to internal libraries, the use of TensorFlow Lite led to a 8 times higher flash memory usage than that of the unquantized reference network, disadvantageous for self-contained prosthetic systems. In response, we offer open-source code solutions that leverage the QKeras platform, effectively reducing flash memory requirements by 24 times compared to Tensorflow Lite. Additionally, we conducted a comprehensive comparison of state-of-the-art microcontrollers. Our results reveal that the adoption of new architectures offers substantial reductions in inference time and power consumption. These improvements pave the way for real-time decoding of motor intent using more advanced machine learning algorithms for daily life usage, possibly enabling more reliable and precise control for prosthetic users.