Terahertz (THz) communications, ranging from 100 GHz to 10 THz, are envisioned as a promising technology for 6G and beyond wireless systems. As foundation of designing THz communications, channel modeling and characterization are crucial to scrutinize the potential of the new spectrum. However, current channel modeling and standardization heavily rely on measurements, which are both time-consuming and costly to obtain in the THz band. Here, we propose a Transfer learning enabled Transformer based Generative Adversarial Network (TT-GAN) for THz channel modeling. Specifically, as a fundamental building block, a GAN is exploited to generate channel parameters, which can substitute measurements. To greatly improve the accuracy, the first T, i.e., a transformer structure with a self-attention mechanism is incorporated in GAN. Still incurring errors compared with ground-truth measurement, the second T, i.e., a transfer learning is designed to solve the mismatch between the formulated network and measurement. The proposed TT-GAN can achieve high accuracy in channel modeling, while requiring only rather limited amount of measurement, which is a promising complementary of channel standardization that fundamentally differs from the current techniques that heavily rely on measurement.

Schizophrenia (SZ) is a complex mental disorder that necessitates accurate and timely diagnosis for effective treatment. Traditional methods for SZ classification often struggle to capture transient EEG features and face high computational complexity. This study proposes a convolutional autoencoder (CAE) to address these challenges by reducing dimensionality and computational complexity. Additionally, we introduce a novel approach utilizing spectral scalograms (SS) combined with EfficientNet (ENB) architectures. The SS, obtained through continuous wavelet transform, reveals temporal and spectral information of EEG signals, aiding in the identification of transient features. ENB models, through transfer learning (TL), extract discriminative features and improve SZ classification accuracy. Experimental evaluation on a comprehensive dataset demonstrates the efficacy of our approach, achieving a five-fold mean cross-validation accuracy of 98.5\% using CAE with a soft voting classifier and 99\% employing SS with the ENB7 model. These results suggest the potential of our methods to enhance SZ diagnosis, surpassing traditional deep learning (DL) and TL techniques. By leveraging CAE and ENBs, this research offers a robust framework for objective SZ classification, promoting early intervention and improved patient outcomes.

The advent of Ultra-Reliable Low Latency Communication (URLLC) alongside the emergence of Open RAN (ORAN) architectures presents unprecedented challenges and opportunities in Radio Resource Management (RRM) for next-generation communication systems. This paper presents a comprehensive trade-off analysis of Deep Reinforcement Learning (DRL) approaches designed to enhance URLLC performance within ORAN's flexible and dynamic framework. By investigating various DRL strategies for optimising RRM parameters, we explore the intricate balance between reliability, latency, and the newfound adaptability afforded by ORAN principles. Through extensive simulation results, our study compares the efficacy of different DRL models in achieving URLLC objectives in an ORAN context, highlighting the potential of DRL to navigate the complexities introduced by ORAN. The proposed study provides valuable insights into the practical implementation of DRL-based RRM solutions in ORAN-enabled wireless networks. It sheds light on the benefits and challenges of integrating DRL and ORAN for URLLC enhancements. Our findings contribute to the ongoing discourse on advancements in URLLC and ORAN, offering a roadmap for future research to pursue efficient, reliable, and flexible communication systems.

We consider a perimeter defense problem in a rooted full tree graph environment in which a single defending vehicle seeks to defend a set of specified vertices, termed as the perimeter from mobile intruders that enter the environment through the tree's leaves. We adopt the technique of competitive analysis to characterize the performance of an online algorithm for the defending vehicle. We first derive fundamental limits on the performance of any online algorithm relative to that of an optimal offline algorithm. Specifically, we give three fundamental conditions for finite, 2, and 3/2 competitive ratios in terms of the environment parameters. We then design and analyze three classes of online algorithms that have provably finite competitiveness under varying environmental parameter regimes. Finally, we give a numerical visualization of these regimes to better show the comparative strengths and weaknesses of each algorithm.

This paper presents the development of a new collaborative road profile estimation and active suspension control framework in connected vehicles, where participating vehicles iteratively refine the road profile estimation and enhance suspension control performance through an iterative learning scheme. Specifically, we develop a robust iterative learning approach to tackle the heterogeneity and model uncertainties in participating vehicles, which are important for practical implementations. In addition, the framework can be adopted as an add-on system to augment existing suspension control schemes. Comprehensive numerical studies are performed to evaluate and validate the proposed framework.

Satellite IoT is an emerging technology that combines the advantages of satellite communication and IoT, providing global coverage, high reliability, and flexible networking. It has a wide range of applications in various fields, including smart agriculture, smart transportation, smart cities, environmental monitoring, and emergency response. With the continuous development of satellite communication, IoT, edge computing, cloud computing, AI, and ML technology, satellite IoT will play an increasingly important role in the future, supporting the digital transformation and sustainable development of society.

Approximate message passing (AMP) algorithms are iterative methods for signal recovery in noisy linear systems. In some scenarios, AMP algorithms need to operate within a distributed network. To address this challenge, the distributed extensions of AMP (D-AMP, FD-AMP) and orthogonal/vector AMP (D-OAMP/D-VAMP) were proposed, but they still inherit the limitations of centralized algorithms. In this letter, we propose distributed memory AMP (D-MAMP) to overcome the IID matrix limitation of D-AMP/FD-AMP, as well as the high complexity and heavy communication cost of D-OAMP/D-VAMP. We introduce a matrix-by-vector variant of MAMP tailored for distributed computing. Leveraging this variant, D-MAMP enables each node to execute computations utilizing locally available observation vectors and transform matrices. Meanwhile, global summations of locally updated results are conducted through message interaction among nodes. For acyclic graphs, D-MAMP converges to the same mean square error performance as the centralized MAMP.

This article introduces the hardware-software co-design of the control system for an autonomy-enabled formula-style high-speed racecar that will be utilized as the deployment platform for high-level autonomy in the first ever head-to-head driverless race called the Indy Autonomous Challenge. The embedded control system needs to facilitate autonomous functionality, including perception, localization, and by-wire actuation, at high speeds and dynamic limits of the vehicle. Rapid maneuvering during the race, however, excites transient dynamics of the vehicle and the actuators. Compared to current autonomous driving focused on highway cruising and urban traffic, transient vehicle control imposes new challenges to the algorithm and system design. The presented work introduces the cascaded control structure employed by the IAC prototype to fully exploit the time scale separation between different control tasks. It is demonstrated by example way how the model-based control strategies and simulation are utilized to inform the decisions in the actuation, computation, perception, and software pipeline design decisions for the first-of-its-kind IAC racecar.

Brain-computer interfaces (BCIs) offer a means to convert neural signals into control signals, providing a potential restoration of movement for people with paralysis. Despite their promise, BCIs face a significant challenge in maintaining decoding accuracy over time due to neural nonstationarities. However, the decoding accuracy of BCI drops severely across days due to the neural data drift. While current recalibration techniques address this issue to a degree, they often fail to leverage the limited labeled data, to consider the signal correlation between two days, or to perform conditional alignment in regression tasks. This paper introduces a novel approach to enhance recalibration performance. We begin with preliminary experiments that reveal the temporal patterns of neural signal changes and identify three critical elements for effective recalibration: global alignment, conditional speed alignment, and feature-label consistency. Building on these insights, we propose the Speed-enhanced Subdomain Adaptation Regression (SSAR) framework, integrating semi-supervised learning with domain adaptation techniques in regression neural decoding. SSAR employs Speed-enhanced Subdomain Alignment (SeSA) for global and speed conditional alignment of similarly labeled data, with Contrastive Consistency Constraint (CCC) to enhance the alignment of SeSA by reinforcing feature-label consistency through contrastive learning. Our comprehensive set of experiments, both qualitative and quantitative, substantiate the superior recalibration performance and robustness of SSAR.

Sensing technology is widely used for comprehending the physical world, with numerous modalities explored in past decades. While there has been considerable work on multi-modality learning, they all require data of all modalities be paired. How to leverage multi-modality data with partially pairings remains an open problem. To tackle this challenge, we introduce the Babel framework, encompassing the neural network architecture, data preparation and processing, as well as the training strategies. Babel serves as a scalable pre-trained multi-modal sensing neural network, currently aligning six sensing modalities, namely Wi-Fi, mmWave, IMU, LiDAR, video, and depth. To overcome the scarcity of complete paired data, the key idea of Babel involves transforming the N-modality alignment into a series of two-modality alignments by devising the expandable network architecture. This concept is also realized via a series of novel techniques, including the pre-trained modality tower that capitalizes on available single-modal networks, and the adaptive training strategy balancing the contribution of the newly incorporated modality with the previously established modality alignment. Evaluation demonstrates Babel's outstanding performance on eight human activity recognition datasets, compared to various baselines e.g., the top multi-modal sensing framework, single-modal sensing networks, and multi-modal large language models. Babel not only effectively fuses multiple available modalities (up to 22% accuracy increase), but also enhance the performance of individual modality (12% averaged accuracy improvement). Case studies also highlight exciting application scenarios empowered by Babel, including cross-modality retrieval (i.e., sensing imaging), and bridging LLM for sensing comprehension.

In clinical applications, X-ray technology is vital for noninvasive examinations like mammography, providing essential anatomical information. However, the radiation risk associated with X-ray procedures raises concerns. X-ray reconstruction is crucial in medical imaging for detailed visual representations of internal structures, aiding diagnosis and treatment without invasive procedures. Recent advancements in deep learning (DL) have shown promise in X-ray reconstruction, but conventional DL methods often require centralized aggregation of large datasets, leading to domain shifts and privacy issues. To address these challenges, we introduce the Hierarchical Framework-based Federated Learning method (HF-Fed) for customized X-ray imaging. HF-Fed tackles X-ray imaging optimization by decomposing the problem into local data adaptation and holistic X-ray imaging. It employs a hospital-specific hierarchical framework and a shared common imaging network called Network of Networks (NoN) to acquire stable features from diverse data distributions. The hierarchical hypernetwork extracts domain-specific hyperparameters, conditioning the NoN for customized X-ray reconstruction. Experimental results demonstrate HF-Fed's competitive performance, offering a promising solution for enhancing X-ray imaging without data sharing. This study significantly contributes to the literature on federated learning in healthcare, providing valuable insights for policymakers and healthcare providers. The source code and pre-trained HF-Fed model are available at \url{https://tisharepo.github.io/Webpage/}.

Immunofluorescent (IF) imaging is crucial for visualizing biomarker expressions, cell morphology and assessing the effects of drug treatments on sub-cellular components. IF imaging needs extra staining process and often requiring cell fixation, therefore it may also introduce artefects and alter endogenouous cell morphology. Some IF stains are expensive or not readily available hence hindering experiments. Recent diffusion models, which synthesise high-fidelity IF images from easy-to-acquire brightfield (BF) images, offer a promising solution but are hindered by training instability and slow inference times due to the noise diffusion process. This paper presents a novel method for the conditional synthesis of IF images directly from BF images along with cell segmentation masks. Our approach employs a Residual Diffusion process that enhances stability and significantly reduces inference time. We performed a critical evaluation against other image-to-image synthesis models, including UNets, GANs, and advanced diffusion models. Our model demonstrates significant improvements in image quality (p<0.05 in MSE, PSNR, and SSIM), inference speed (26 times faster than competing diffusion models), and accurate segmentation results for both nuclei and cell bodies (0.77 and 0.63 mean IOU for nuclei and cell true positives, respectively). This paper is a substantial advancement in the field, providing robust and efficient tools for cell image analysis.

The human brain has the capability to associate the unknown person's voice and face by leveraging their general relationship, referred to as ``cross-modal speaker verification''. This task poses significant challenges due to the complex relationship between the modalities. In this paper, we propose a ``Multi-stage Face-voice Association Learning with Keynote Speaker Diarization''~(MFV-KSD) framework. MFV-KSD contains a keynote speaker diarization front-end to effectively address the noisy speech inputs issue. To balance and enhance the intra-modal feature learning and inter-modal correlation understanding, MFV-KSD utilizes a novel three-stage training strategy. Our experimental results demonstrated robust performance, achieving the first rank in the 2024 Face-voice Association in Multilingual Environments (FAME) challenge with an overall Equal Error Rate (EER) of 19.9%. Details can be found in https://github.com/TaoRuijie/MFV-KSD.

The recent rise of EEG-based end-to-end deep learning models presents a significant challenge in elucidating how these models process raw EEG signals and generate predictions in the frequency domain. This challenge limits the transparency and credibility of EEG-based end-to-end models, hindering their application in security-sensitive areas. To address this issue, we propose a mask perturbation method to explain the behavior of end-to-end models in the frequency domain. Considering the characteristics of EEG data, we introduce a target alignment loss to mitigate the out-of-distribution problem associated with perturbation operations. Additionally, we develop a perturbation generator to define perturbation generation in the frequency domain. Our explanation method is validated through experiments on multiple representative end-to-end deep learning models in the EEG decoding field, using an established EEG benchmark dataset. The results demonstrate the effectiveness and superiority of our method, and highlight its potential to advance research in EEG-based end-to-end models.

Reconfigurable Intelligent Surface (RIS) technology has gained significant attention due to its ability to enhance the performance of wireless communication systems. The main advantage of RIS is that it can be strategically placed in the environment to control wireless signals, enabling improvements in coverage, capacity, and energy efficiency. In this paper, we investigate a scenario in which a drone, equipped with a RIS, travels from an initial point to a target destination. In this scenario, the aerial RIS (ARIS) is deployed to establish a direct link between the base station and obstructed users. Our objective is to maximize the energy efficiency of the ARIS while taking into account its dynamic model including its velocity and acceleration along with the phase shift of the RIS. To this end, we formulate the energy efficiency problem under the constraints of the dynamic model of the drone. The studied problem is challenging to solve. To address this, we proceed as follows. First, we introduce an efficient solution that involves decoupling the phase shift optimization and the trajectory design. Specifically, the closed-form expression of the phase-shift is obtained using a convex approximation, which is subsequently integrated into the trajectory design problem. We then employ tools inspired by economic model predictive control (EMPC) to solve the resulting trajectory optimization. Our simulation results show a significant improvement in energy efficiency against the scenario where the dynamic model of the UAV is ignored.

Inverse problems, such as accelerated MRI reconstruction, are ill-posed and an infinite amount of possible and plausible solutions exist. This may not only lead to uncertainty in the reconstructed image but also in downstream tasks such as semantic segmentation. This uncertainty, however, is mostly not analyzed in the literature, even though probabilistic reconstruction models are commonly used. These models can be prone to ignore plausible but unlikely solutions like rare pathologies. Building on MRI reconstruction approaches based on diffusion models, we add guidance to the diffusion process during inference, generating two meaningfully diverse reconstructions corresponding to an upper and lower bound segmentation. The reconstruction uncertainty can then be quantified by the difference between these bounds, which we coin the 'uncertainty boundary'. We analyzed the behavior of the upper and lower bound segmentations for a wide range of acceleration factors and found the uncertainty boundary to be both more reliable and more accurate compared to repeated sampling. Code is available at https://github.com/NikolasMorshuis/SGR

Backscatter communication (BC) has emerged as a key technology to satisfy the increasing need for low-cost and green Internet-of-Things (IoT) connectivity, especially in large-scale deployments. Unlike the monostatic BC (MoBC), the bistatic BC (BiBC) has the possibility to decrease the round-trip path loss by having the carrier emitter (CE) and the reader in different locations. Therefore, this work investigates the BiBC in the context of cell-free multiple-input multiple-output (MIMO) networks by exploring the optimal selection of CE and reader among all access points, leveraging prior knowledge about the area where the backscatter device (BD) is located. First, a maximum a posteriori probability (MAP) detector to decode the BD information bits is derived. Then, the exact probability of error for this detector is obtained. In addition, an algorithm to select the best CE-reader pair for serving the specified area is proposed. Finally, simulation results show that the error performance of the BC is improved by the proposed algorithm compared to the benchmark scenario.

The segmentation of cell nuclei in tissue images stained with the blood dye hematoxylin and eosin (H$\&$E) is essential for various clinical applications and analyses. Due to the complex characteristics of cellular morphology, a large receptive field is considered crucial for generating high-quality segmentation. However, previous methods face challenges in achieving a balance between the receptive field and computational burden. To address this issue, we propose LKCell, a high-accuracy and efficient cell segmentation method. Its core insight lies in unleashing the potential of large convolution kernels to achieve computationally efficient large receptive fields. Specifically, (1) We transfer pre-trained large convolution kernel models to the medical domain for the first time, demonstrating their effectiveness in cell segmentation. (2) We analyze the redundancy of previous methods and design a new segmentation decoder based on large convolution kernels. It achieves higher performance while significantly reducing the number of parameters. We evaluate our method on the most challenging benchmark and achieve state-of-the-art results (0.5080 mPQ) in cell nuclei instance segmentation with only 21.6% FLOPs compared with the previous leading method. Our source code and models are available at https://github.com/hustvl/LKCell.

Deep learning, especially convolutional neural networks (CNNs) and Transformer architectures, have become the focus of extensive research in medical image segmentation, achieving impressive results. However, CNNs come with inductive biases that limit their effectiveness in more complex, varied segmentation scenarios. Conversely, while Transformer-based methods excel at capturing global and long-range semantic details, they suffer from high computational demands. In this study, we propose CSWin-UNet, a novel U-shaped segmentation method that incorporates the CSWin self-attention mechanism into the UNet to facilitate horizontal and vertical stripes self-attention. This method significantly enhances both computational efficiency and receptive field interactions. Additionally, our innovative decoder utilizes a content-aware reassembly operator that strategically reassembles features, guided by predicted kernels, for precise image resolution restoration. Our extensive empirical evaluations on diverse datasets, including synapse multi-organ CT, cardiac MRI, and skin lesions, demonstrate that CSWin-UNet maintains low model complexity while delivering high segmentation accuracy.

The Antillean manatee (\emph{Trichechus manatus}) is an endangered herbivorous aquatic mammal whose role as an ecological balancer and umbrella species underscores the importance of its conservation. An innovative approach to monitor manatee populations is passive acoustic monitoring (PAM), where vocalisations are extracted from submarine audio. We propose a novel end-to-end approach to detect manatee vocalisations building on the Audio Spectrogram Transformer (AST). In a transfer learning spirit, we fine-tune AST to detect manatee calls by redesigning its filterbanks and adapting a real-world dataset containing partial positive labels. Our experimental evaluation reveals the two key features of the proposed model: i) it performs on par with the state of the art without requiring hand-tuned denoising or detection stages, and ii) it can successfully identify missed vocalisations in the training dataset, thus reducing the workload of expert bioacoustic labellers. This work is a preliminary relevant step to develop novel, user-friendly tools for the conservation of the different species of manatees.

This paper investigates the estimation problem of the pose (orientation and position) and linear velocity of a rigid body, as well as the landmark positions, using an inertial measurement unit (IMU) and a monocular camera. First, we propose a globally exponentially stable (GES) linear time-varying (LTV) observer for the estimation of body-frame landmark positions and velocity, using IMU and monocular bearing measurements. Thereafter, using the gyro measurements, some landmarks known in the inertial frame and the estimates from the LTV observer, we propose a nonlinear pose observer on $\SO(3)\times \mathbb{R}^3$. The overall estimation system is shown to be almost globally asymptotically stable (AGAS) using the notion of almost global input-to-state stability (ISS). Interestingly, we show that with the knowledge (in the inertial frame) of a small number of landmarks, we can recover (under some conditions) the unknown positions (in the inertial frame) of a large number of landmarks. Numerical simulation results are presented to illustrate the performance of the proposed estimation scheme.

Computer vision models are increasingly capable of classifying ovarian epithelial cancer subtypes, but they differ from pathologists by processing small tissue patches at a single resolution. Multi-resolution graph models leverage the spatial relationships of patches at multiple magnifications, learning the context for each patch. In this study, we conduct the most thorough validation of a graph model for ovarian cancer subtyping to date. Seven models were tuned and trained using five-fold cross-validation on a set of 1864 whole slide images (WSIs) from 434 patients treated at Leeds Teaching Hospitals NHS Trust. The cross-validation models were ensembled and evaluated using a balanced hold-out test set of 100 WSIs from 30 patients, and an external validation set of 80 WSIs from 80 patients in the Transcanadian Study. The best-performing model, a graph model using 10x+20x magnification data, gave balanced accuracies of 73%, 88%, and 99% in cross-validation, hold-out testing, and external validation, respectively. However, this only exceeded the performance of attention-based multiple instance learning in external validation, with a 93% balanced accuracy. Graph models benefitted greatly from using the UNI foundation model rather than an ImageNet-pretrained ResNet50 for feature extraction, with this having a much greater effect on performance than changing the subsequent classification approach. The accuracy of the combined foundation model and multi-resolution graph network offers a step towards the clinical applicability of these models, with a new highest-reported performance for this task, though further validations are still required to ensure the robustness and usability of the models.

In smart city development, the automatic detection of structures and vehicles within urban or suburban areas via array radar (airborne or vehicle platforms) becomes crucial. However, the inescapable multipath effect adversely affects the radar's capability to detect and track targets. Frequency Diversity Array (FDA)-MIMO radar offers innovative solutions in mitigating multipath due to its frequency flexibility and waveform diversity traits amongst array elements. Hence, utilizing FDA-MIMO radar, this research proposes a multipath discrimination and suppression strategy to augment target detection and suppress false alarms. The primary advancement is the transformation of conventional multipath suppression into a multipath recognition issue, thereby enabling multipath components from single-frame echo data to be separated without prior knowledge. By offsetting the distance steering vectors of different objects to be detected, the accurate spectral information corresponding to the current distance unit can be extracted during spatial spectrum estimation. The direct and multipath components are differentiated depending on whether the transmitting and receiving angles match. Additionally, to mitigate high-order multipath, the echo intensity of multipath components is reduced via joint optimization of array transmit weighting and frequency increment. The numerical results show that the proposed algorithm can identify multipath at different distances in both single-target and multi-target scenarios, which is superior to the general MIMO radar.

Throughout human history, epidemics have been a constant presence. Understanding their dynamics is essential to predict scenarios and make substantiated decisions. Mathematical models are powerful tools to describe an epidemic behavior. Among the most used, the compartmental ones stand out, dividing population into classes with well-defined characteristics. One of the most known is the $SIR$ model, based on a set of differential equations describing the rates of change of three categories over time. These equations take into account parameters such as the disease transmission rate and the recovery rate, which both change over time. However, classical models use constant parameters and can not describe the behavior of a disease over long periods. In this work, it is proposed a $SIR$ model with time-varying transmission rate parameter with a method to estimate this parameter based on an optimization problem, which minimizes the sum of the squares of the errors between the model and historical data. Additionally, based on the infection rates determined by the algorithm, the model's ability to predict disease activity in future scenarios was also investigated. Epidemic data released by the government of the State of Rio Grande do Sul in Brazil was used to evaluate the models, where the models shown a very good forecasting ability, resulting in errors for predicting the total number of accumulated infected persons of 0.13% for 7 days ahead and 0.6% for 14 days ahead.

We consider optimal swarm control problems where two different classes of agents are present. Continuum idealizations of large-scale swarms are used where the dynamics describe the evolution of the spatially-distributed densities of each agent class. The problem formulation we adopt is motivated by applications where agents of one class are assigned to agents of the other class, which we refer to as demand and resource agents respectively. Assignments have costs related to the distances between mutually assigned agents, and the overall cost of an assignment is quantified by a Wasserstein distance between the densities of the two agent classes. When agents can move, the assignment cost can decrease at the expense of a physical motion cost, and this tradeoff sets up a nonlinear, infinite-dimensional optimal control problem. We show that in one spatial dimension, this problem can be converted to an infinite-dimensional, but decoupled, linear-quadratic (LQ) tracking problem when expressed in terms of the respective quantile functions. Solutions are given in the general one-dimensional case, as well as in the special cases of constant and periodically time-varying demands.

In this paper, a Full Duplex (FD) eXtremely Large (XL) Multiple-Input Multiple-Output (MIMO) node equipped with reconfigurable metasurface antennas at its transmission and reception sides is considered, which is optimized for simultaneous multi-user communications and sensing in the near-field regime at THz frequencies. We first present a novel Position Error Bound (PEB) analysis for the spatial parameters of multiple targets in the vicinity of the FD node, via the received backscattered data signals, and devise an optimization framework for its metasurface-based precoder and combiner. Then, we formulate and solve an optimization problem aiming at the downlink sum-rate maximization, while simultaneously ensuring a minimum PEB requirement for targets' localization. Our simulation results for a sub-THz system setup validate the joint near-field communications and sensing capability of the proposed FD XL MIMO scheme with metasurfaces antennas, showcasing the interplay of its various design parameters.

Helmholtz stereopsis is one the versatile techniques for 3D geometry reconstruction from 2D images of objects with unknown and arbitrary reflectance surfaces. HS eliminates the need for surface reflectance, a challenging parameter to measure, based on the Helmholtz reciprocity principle. Its Bayesian formulation using maximum a posteriori (MAP) probability approach has significantly improved reconstruction accuracy of HS method. This framework enables the inclusion of smoothness priors which enforces observations and neighborhood information in the formulation. We used Markov Random Fields (MRF) which is a powerful tool to integrate diverse prior contextual information and solved the MAP-MRF using belief propagation algorithm. We propose a new smoothness function utilizing the normal field integration method for refined depth estimation within the Bayesian framework. Utilizing three pairs of images with different viewpoints, our approach demonstrates superior depth label accuracy compared to conventional Bayesian methods. Experimental results indicate that our proposed method yields a better depth map with reduced RMS error, showcasing its efficacy in improving depth estimation within Helmholtz stereopsis.

In this paper, we present Reshape Dimensions Network (ReDimNet), a novel neural network architecture for extracting utterance-level speaker representations. Our approach leverages dimensionality reshaping of 2D feature maps to 1D signal representation and vice versa, enabling the joint usage of 1D and 2D blocks. We propose an original network topology that preserves the volume of channel-timestep-frequency outputs of 1D and 2D blocks, facilitating efficient residual feature maps aggregation. Moreover, ReDimNet is efficiently scalable, and we introduce a range of model sizes, varying from 1 to 15 M parameters and from 0.5 to 20 GMACs. Our experimental results demonstrate that ReDimNet achieves state-of-the-art performance in speaker recognition while reducing computational complexity and the number of model parameters.

In this article we present the biological effect of antenna topology on a users body. At different values of exposed frequency, the absorbent nature varies in human body. One of the major factors to be taken into consideration for designing 6G mobile antenna is the biological effect and Electromagnetic Field Exposure (EMF).

The 5G New Radio NR technology is under standardization process by 3GPP to provide outline for a new radio interface for the next generation of cellular networks. The aim of the 5G networks include not only to provide enhanced capacity coverage but also support advanced services such as enhanced mobile broadband (eMBB) Ultra-Reliable Low Latency Communication URLLC massive Machine Type Communication mMTC. Key features of NR include Ultra lean carrier design to minimize the power consumption by limiting the always-on signal transmissions and to reduce interference in the neighboring cells . Another feature is the use of massive number of antennas for transmission as well as reception of signals. This rise in the number of antennas to provide a greater coverage brings about various challenges and impact in the system. With the increase in investigations in the mmWave frequencies, there is a need to investigate the health hazards they have on human body and the environment at large. This paper intends to provide an insight into the harmful impacts of Radio Frequency RF fields. The radiation metric to study the RF impact for far field is power density and for near field is Specific Absorption Rate SAR. These are the two main EM radiation metrics to find out the exposure due to uplink and downlink phenomenon in mobile communications. Mobile communication systems are addressed particularly to discuss the Electromagnetic EM Radiation impact as smart phones are used in close proximity to the body. A proposal in the form of Thermal Radiation TR mode is given to reduce the radiations emitted from a mobile phone. The performance of the proposed mode is validated from the results by achieving reduced power density, complexity and exposure ratio.

With the wireless internet access being increasingly popular with services such as HD video streaming and so on, the demand for high data consuming applications is also rising. This increment in demand is coupled with a proportional rise in the power consumption. It is required that the internet traffic is offloaded to technologies that serve the users and contribute in energy consumption. There is a need to decrease the carbon footprint in the atmosphere and also make the network safe and reliable. In this article we propose a hybrid system of RF (Radio Frequency) and VLC (Visible Light Communication) for indoor communication that can provide communication along with illumination with least power consumption. The hybrid network is viable as it utilizes power with respect to the user demand and maintains the required Quality of ServiceQoS and Quality of Experience QoE for a particular application in use. This scheme aims for Green Communication and reduction in Electromagnetic EM Radiation. A comparative analysis for RF communication, Hybrid RF+ VLC and pure VLC is made and simulations are carried out using Python, Scilab and MathWorks tool. The proposal achieves high energy efficiency of about 37% low Specific Absorption Rate (SAR) lower incident and absorbed power density complexity and temperature elevation in human body tissues exposed to the radiation. It also enhances the battery lifetime of the mobile device in use by increasing the lifetime approximately by 7 hours as validated from the obtained results. Thus the overall network reliability and safety factor is enhanced with the proposed approach.

The graph linear canonical transform (GLCT) is presented as an extension of the graph Fourier transform (GFT) and the graph fractional Fourier transform (GFrFT), offering more flexibility as an effective tool for graph signal processing. In this paper, we introduce a GLCT based on chirp multiplication-chirp convolution-chirp multiplication decomposition (CM-CC-CM-GLCT), which irrelevant to sampling periods and without oversampling operation. Various properties and special cases of the CM-CC-CM-GLCT are derived and discussed. In terms of computational complexity, additivity, and reversibility, we compare the CM-CC-CM-GLCT and the GLCT based on the central discrete dilated Hermite function (CDDHFs-GLCT). Theoretical analysis demonstrates that the computational complexity of the CM-CC-CM-GLCT is significantly reduced. Simulation results indicate that the CM-CC-CM-GLCT achieves similar additivity to the CDDHFs-GLCT. Notably, the CM-CC-CM-GLCT exhibits better reversibility.

Deep learning models have become a critical tool for analysis and classification of musical data. These models operate either on the audio signal, e.g. waveform or spectrogram, or on a symbolic representation, such as MIDI. In the latter, musical information is often reduced to basic features, i.e. durations, pitches and velocities. Most existing works then rely on generic tokenization strategies from classical natural language processing, or matrix representations, e.g. piano roll. In this work, we evaluate how enriched representations of symbolic data can impact deep models, i.e. Transformers and RNN, for music style classification. In particular, we examine representations that explicitly incorporate musical information implicitly present in MIDI-like encodings, such as rhythmic organization, and show that they outperform generic tokenization strategies. We introduce a new tree-based representation of MIDI data built upon a context-free musical grammar. We show that this grammar representation accurately encodes high-level rhythmic information and outperforms existing encodings on the GrooveMIDI Dataset for drumming style classification, while being more compact and parameter-efficient.

We present a modular approach to building cascade speech translation (AST) models that guarantees that the resulting model performs no worse than the 1-best cascade baseline while preserving state-of-the-art speech recognition (ASR) and text translation (MT) performance for a given task. Our novel contribution is the use of an ``exporter'' layer that is trained under L2-loss to ensure a strong match between ASR embeddings and the MT token embeddings for the 1-best sequence. The ``exporter'' output embeddings are fed directly to the MT model in lieu of 1-best token embeddings, thus guaranteeing that the resulting model performs no worse than the 1-best cascade baseline, while allowing back-propagation gradient to flow from the MT model into the ASR components. The matched-embeddings cascade architecture provide a significant improvement over its 1-best counterpart in scenarios where incremental training of the MT model is not an option and yet we seek to improve quality by leveraging (speech, transcription, translated transcription) data provided with the AST task. The gain disappears when the MT model is incrementally trained on the parallel text data available with the AST task. The approach holds promise for other scenarios that seek to couple ASR encoders and immutable text models, such at large language models (LLM).

Narrowband Internet of Things (NB-IoT) is a promising technology designated specially by the 3rd Generation Partnership Project (3GPP) to meet the growing demand of massive machine-type communications (mMTC). More and more industrial companies choose NB-IoT network as the solution to mMTC due to its unique design and technical specification released by 3GPP. In order to evaluate the performance of NB-IoT network, we design a system-level simulation for NB-IoT network in this paper. In particular, the structure of system-level simulator are divided into four parts, i.e., initialization, pre-generation, main simulation loop and post-processing. Moreover, three key techniques are developed in the implementation of NB-IoT network by accounting for enhanced coverage, massive connection and low-power consumption. Simulation results demonstrate the cumulative distribution function curves of signal-to-interference-and-noise ratio are fully compliant with industrial standard, and the performance of throughput explains how NB-IoT network realize massive connection at the cost of data rate.

Speech emotion recognition (SER) systems often struggle in real-world environments, where ambient noise severely degrades their performance. This paper explores a novel approach that exploits prior knowledge of testing environments to maximize SER performance under noisy conditions. To address this task, we propose a text-guided, environment-aware training where an SER model is trained with contaminated speech samples and their paired noise description. We use a pre-trained text encoder to extract the text-based environment embedding and then fuse it to a transformer-based SER model during training and inference. We demonstrate the effectiveness of our approach through our experiment with the MSP-Podcast corpus and real-world additive noise samples collected from the Freesound repository. Our experiment indicates that the text-based environment descriptions processed by a large language model (LLM) produce representations that improve the noise-robustness of the SER system. In addition, our proposed approach with an LLM yields better performance than our environment-agnostic baselines, especially in low signal-to-noise ratio (SNR) conditions. When testing at -5dB SNR level, our proposed method shows better performance than our best baseline model by 31.8 % (arousal), 23.5% (dominance), and 9.5% (valence).

Movable antenna (MA) technology can flexibly reconfigure wireless channels by adjusting antenna positions in a local region, thus owing great potential for enhancing communication performance. This letter investigates MA technology enabled multiuser uplink communications over general Rician fading channels, which consist of a base station (BS) equipped with the MA array and multiple single-antenna users. Since it is practically challenging to collect all instantaneous channel state information (CSI) by traversing all possible antenna positions at the BS, we instead propose a two-timescale scheme for maximizing the ergodic sum rate. Specifically, antenna positions at the BS are first optimized using only the statistical CSI. Subsequently, the receiving beamforming at the BS (for which we consider the three typical zero-forcing (ZF), minimum mean-square error (MMSE) and MMSE with successive interference cancellation (MMSE-SIC) receivers) is designed based on the instantaneous CSI with optimized antenna positions, thus significantly reducing practical implementation complexities. The formulated problems are highly non-convex and we develop projected gradient ascent (PGA) algorithms to effectively handle them. Simulation results illustrate that compared to conventional fixed-position antenna (FPA) array, the MA array can achieve significant performance gains by reaping an additional spatial degree of freedom.

Parkinson's disease (PD), the second most prevalent neurodegenerative disorder worldwide, frequently presents with early-stage speech impairments. Recent advancements in Artificial Intelligence (AI), particularly deep learning (DL), have significantly enhanced PD diagnosis through the analysis of speech data. Nevertheless, the progress of research is restricted by the limited availability of publicly accessible speech-based PD datasets, primarily due to privacy and ethical concerns. This review covers the latest DL-based AI approaches for speech-based PD classification, focusing on performance, available resources and associated challenges of 33 scientific works published between 2020 and March 2024. These DL approaches are categorized into end-to-end (E2E) learning, transfer learning (TL) and deep acoustic features (DAF) extraction. Among E2E approaches, Convolutional Neural Networks (CNNs) are prevalent, though Transformers are increasingly popular. E2E approaches face challenges such as limited data and computational resources, especially with Transformers. TL addresses these issues by providing more robust PD diagnosis and better generalizability across languages. DAF extraction aims to improve the explainability and interpretability of results by examining the specific effects of deep features on both other DL approaches and more traditional machine learning (ML) methods. However, it often underperforms compared to E2E and TL approaches. This review also discusses unresolved issues related to bias, explainability and privacy, highlighting the need for future research.

Background: Benchmarking medical decision support algorithms often struggles due to limited access to datasets, narrow prediction tasks, and restricted input modalities. These limitations affect their clinical relevance and performance in high-stakes areas like emergency care, complicating replication, validation, and improvement of benchmarks. Methods: We introduce a dataset based on MIMIC-IV, benchmarking protocol, and initial results for evaluating multimodal decision support in the emergency department (ED). We use diverse data modalities from the first 1.5 hours of patient arrival, including demographics, biometrics, vital signs, lab values, and electrocardiogram waveforms. We analyze 1443 clinical labels across two contexts: predicting diagnoses with ICD-10 codes and forecasting patient deterioration. Results: Our multimodal diagnostic model achieves an AUROC score over 0.8 in a statistically significant manner for 357 out of 1428 conditions, including cardiac issues like myocardial infarction and non-cardiac conditions such as renal disease and diabetes. The deterioration model scores above 0.8 in a statistically significant manner for 13 out of 15 targets, including critical events like cardiac arrest and mechanical ventilation, ICU admission as well as short- and long-term mortality. Incorporating raw waveform data significantly improves model performance, which represents one of the first robust demonstrations of this effect. Conclusions: This study highlights the uniqueness of our dataset, which encompasses a wide range of clinical tasks and utilizes a comprehensive set of features collected early during the emergency after arriving at the ED. The strong performance, as evidenced by high AUROC scores across diagnostic and deterioration targets, underscores the potential of our approach to revolutionize decision-making in acute and emergency medicine.

Time-critical tasks such as drone racing typically cover large operation areas. However, it is difficult and computationally intensive for current time-optimal motion planners to accommodate long flight distances since a large yet unknown number of knot points is required to represent the trajectory. We present a polynomial-based automatic optimal synthesis (AOS) approach that can address this challenge. Our method not only achieves superior time optimality but also maintains a consistently low computational cost across different ranges while considering the full quadrotor dynamics. First, we analyze the properties of time-optimal quadrotor maneuvers to determine the minimal number of polynomial pieces required to capture the dominant structure of time-optimal trajectories. This enables us to represent substantially long minimum-time trajectories with a minimal set of variables. Then, a robust optimization scheme is developed to handle arbitrary start and end conditions as well as intermediate waypoints. Extensive comparisons show that our approach is faster than the state-of-the-art approach by orders of magnitude with comparable time optimality. Real-world experiments further validate the quality of the resulting trajectories, demonstrating aggressive time-optimal maneuvers with a peak velocity of 8.86 m/s.

In this work we evaluate the utility of synthetic data for training automatic speech recognition (ASR). We use the ASR training data to train a text-to-speech (TTS) system similar to FastSpeech-2. With this TTS we reproduce the original training data, training ASR systems solely on synthetic data. For ASR, we use three different architectures, attention-based encoder-decoder, hybrid deep neural network hidden Markov model and a Gaussian mixture hidden Markov model, showing the different sensitivity of the models to synthetic data generation. In order to extend previous work, we present a number of ablation studies on the effectiveness of synthetic vs. real training data for ASR. In particular we focus on how the gap between training on synthetic and real data changes by varying the speaker embedding or by scaling the model size. For the latter we show that the TTS models generalize well, even when training scores indicate overfitting.

Federated Learning (FL) is the most widely adopted collaborative learning approach for training decentralized Machine Learning (ML) models by exchanging learning between clients without sharing the data and compromising privacy. However, since great data similarity or homogeneity is taken for granted in all FL tasks, FL is still not specifically designed for the industrial setting. Rarely this is the case in industrial data because there are differences in machine type, firmware version, operational conditions, environmental factors, and hence, data distribution. Albeit its popularity, it has been observed that FL performance degrades if the clients have heterogeneous data distributions. Therefore, we propose a Lightweight Industrial Cohorted FL (LICFL) algorithm that uses model parameters for cohorting without any additional on-edge (clientlevel) computations and communications than standard FL and mitigates the shortcomings from data heterogeneity in industrial applications. Our approach enhances client-level model performance by allowing them to collaborate with similar clients and train more specialized or personalized models. Also, we propose an adaptive aggregation algorithm that extends the LICFL to Adaptive LICFL (ALICFL) for further improving the global model performance and speeding up the convergence. Through numerical experiments on real-time data, we demonstrate the efficacy of the proposed algorithms and compare the performance with existing approaches.

In the event of a total loss of thrust a pilot must identify a reachable landing site and subsequently execute a forced landing. To do so this, they must estimate which region on the ground can be reached safely in gliding flight. We call this the gliding reachable region (GRR). To compute the GRR, we employ an optimal control formulation aiming to reach a point in space while minimizing altitude loss. A simplified model of the aircraft's dynamics is used, where the effect of turns is neglected. The resulting equations are discretized on a grid and solved numerically. Our algorithm for computing the GRR is fast enough to run in real time during flight, it accounts for ground obstacles and wind, and for each point in the GRR it outputs the path to reach it with minimal loss of altitude. A related problem is estimating the minimal altitude an aircraft needs to glide to a given airfield in the presence of obstacles. This information enables pilots to plan routes that always have an airport within gliding distance. We formalize this problem using an optimal control formulation based on the same aircraft dynamics model. The resulting equations are solved with a second algorithm that outputs the minimal re-entry altitude and the paths to reach the airfield from any position while avoiding obstacles. The algorithms we develop are based on the Ordered Upwind Method and the Fast Marching Method.

Music two-tower multimodal systems integrate audio and text modalities into a joint audio-text space, enabling direct comparison between songs and their corresponding labels. These systems enable new approaches for classification and retrieval, leveraging both modalities. Despite the promising results they have shown for zero-shot classification and retrieval tasks, closer inspection of the embeddings is needed. This paper evaluates the inherent zero-shot properties of joint audio-text spaces for the case-study of instrument recognition. We present an evaluation and analysis of two-tower systems for zero-shot instrument recognition and a detailed analysis of the properties of the pre-joint and joint embeddings spaces. Our findings suggest that audio encoders alone demonstrate good quality, while challenges remain within the text encoder or joint space projection. Specifically, two-tower systems exhibit sensitivity towards specific words, favoring generic prompts over musically informed ones. Despite the large size of textual encoders, they do not yet leverage additional textual context or infer instruments accurately from their descriptions. Lastly, a novel approach for quantifying the semantic meaningfulness of the textual space leveraging an instrument ontology is proposed. This method reveals deficiencies in the systems' understanding of instruments and provides evidence of the need for fine-tuning text encoders on musical data.

Recent literature uses language to build foundation models for audio. These Audio-Language Models (ALMs) are trained on a vast number of audio-text pairs and show remarkable performance in tasks including Text-to-Audio Retrieval, Captioning, and Question Answering. However, their ability to engage in more complex open-ended tasks, like Interactive Question-Answering, requires proficiency in logical reasoning -- a skill not yet benchmarked. We introduce the novel task of Audio Entailment to evaluate an ALM's deductive reasoning ability. This task assesses whether a text description (hypothesis) of audio content can be deduced from an audio recording (premise), with potential conclusions being entailment, neutral, or contradiction, depending on the sufficiency of the evidence. We create two datasets for this task with audio recordings sourced from two audio captioning datasets -- AudioCaps and Clotho -- and hypotheses generated using Large Language Models (LLMs). We benchmark state-of-the-art ALMs and find deficiencies in logical reasoning with both zero-shot and linear probe evaluations. Finally, we propose "caption-before-reason", an intermediate step of captioning that improves the zero-shot and linear-probe performance of ALMs by an absolute 6% and 3%, respectively.

In this work, we present composite Bernstein polynomials as a direct collocation method for approximating optimal control problems. An analysis of the convergence properties of composite Bernstein polynomials is provided, and beneficial properties of composite Bernstein polynomials for the solution of optimal control problems are discussed. The efficacy of the proposed approximation method is demonstrated through a bang-bang example. Lastly, we apply this method to a motion planning problem, offering a practical solution that emphasizes the ability of this method to solve complex optimal control problems.

Earthquakes are commonly estimated using physical seismic stations, however, due to the installation requirements and costs of these stations, global coverage quickly becomes impractical. An efficient and lower-cost alternative is to develop machine learning models to globally monitor earth observation data to pinpoint regions impacted by these natural disasters. However, due to the small amount of historically recorded earthquakes, this becomes a low-data regime problem requiring algorithmic improvements to achieve peak performance when learning to regress earthquake magnitude. In this paper, we propose to pose the estimation of earthquake magnitudes as a metric-learning problem, training models to not only estimate earthquake magnitude from Sentinel-1 satellite imagery but to additionally rank pairwise samples. Our experiments show at max a 30%+ improvement in MAE over prior regression-only based methods, particularly transformer-based architectures.

Integrating cameras into wireless smart rings has been challenging due to size and power constraints. We introduce IRIS, the first wireless vision-enabled smart ring system for smart home interactions. Equipped with a camera, Bluetooth radio, inertial measurement unit (IMU), and an onboard battery, IRIS meets the small size, weight, and power (SWaP) requirements for ring devices. IRIS is context-aware, adapting its gesture set to the detected device, and can last for 16-24 hours on a single charge. IRIS leverages the scene semantics to achieve instance-level device recognition. In a study involving 23 participants, IRIS consistently outpaced voice commands, with a higher proportion of participants expressing a preference for IRIS over voice commands regarding toggling a device's state, granular control, and social acceptability. Our work pushes the boundary of what is possible with ring form-factor devices, addressing system challenges and opening up novel interaction capabilities.

This paper investigates federated learning (FL) in a multi-hop communication setup, such as in constellations with inter-satellite links. In this setup, part of the FL clients are responsible for forwarding other client's results to the parameter server. Instead of using conventional routing, the communication efficiency can be improved significantly by using in-network model aggregation at each intermediate hop, known as incremental aggregation (IA). Prior works [1] have indicated diminishing gains for IA under gradient sparsification. Here we study this issue and propose several novel correlated sparsification methods for IA. Numerical results show that, for some of these algorithms, the full potential of IA is still available under sparsification without impairing convergence. We demonstrate a 15x improvement in communication efficiency over conventional routing and a 11x improvement over state-of-the-art (SoA) sparse IA.