The rapid evolution and convergence of wireless technologies and vertical applications have fundamentally reshaped our lifestyles and industries. Future wireless networks, especially 6G, are poised to support a wide range of applications enabled by heterogeneous services, leveraging both traditional connectivity-centric functions and emerging beyond-communication capabilities, particularly localization, sensing, and synchronization. However, integrating these new capabilities into a unified 6G paradigm presents significant challenges. This article provides an in-depth analysis of these technical challenges for integrative 6G design and proposes three strategies for concurrent heterogeneous service provisioning, with the aggregated goal of maximizing integration gains while minimizing service provisioning overhead. First, we adopt multi-dimensional multiple access (MDMA) as an inclusive enabling platform to flexibly integrate various capabilities by shared access to multidimensional radio resources. Next, we propose value-oriented heterogeneous service provisioning to maximize the integration gain through situation-aware MDMA. To enhance scalability, we optimize control and user planes by eliminating redundant control information and enabling service-oriented prioritization. Finally, we evaluate the proposed framework with a case study on integrated synchronization and communication, demonstrating its potential for concurrent heterogeneous service provisioning.

Objective: The interpretation of electrocardiogram (ECG) signals is vital for the diagnosis of cardiac conditions.Traditional methods rely on expert knowledge, which is time consuming, costly and potentially misses subtle features. Deep learning has shown promise in ECG interpretation, but model transparency is often overlooked in literature but desired in medical contexts. Methods: We introduce an explainable AI method for ECG classification by partitioning the latent space of a variational autoencoder (VAE) into a label-specific and a non-label-specific subset. By optimizing both subsets for signal reconstruction and one subset also for prediction while constraining the other from learning label-specific information with an adversarial network, the latent space is disentangled in a supervised manner. This latent space is subsequently leveraged to create enhanced visualizations for ECG feature interpretation by means of attribute manipulation. As a proof of concept, we predict the left ventricular function (LVF), a critical prognostic determinant in cardiac disease, from the ECG. Results: Our study demonstrates the effective segregation of LVF-specific information within a single dimension of the VAE latent space, without compromising classification performance. We show that the proposed model improves state-of-the-art VAE methods (AUC 0.832 vs. 0.790, F1 0.688 vs. 0.640) in prediction and performs comparable to ground truth LVF (concordance 0.72 vs.0.72) in predicting survival. Conclusion: The model facilitates the interpretation of LVF predictions by providing visual context to the ECG signals, offering a general explainable and predictive AI method. Significance: Our explainable AI model can potentially reduce time and expertise required for ECG analysis.

With the growing demand for global connectivity and the proliferation of Internet of Things (IoT) devices, nonterrestrial networks (NTNs) have become critical in providing seamless and ubiquitous coverage and connectivity. However, integrating different NTN platforms and their integration with terrestrial networks presents significant challenges, such as managing interference and optimizing resource allocation. This work proposes a transmissive reconfigurable intelligent surface (T-RIS)-equipped low earth orbit (LEO) satellite communication in cognitive radio-enabled integrated NTNs. In the proposed system, a geostationary (GEO) satellite operates as a primary network, and a T-RIS-equipped LEO satellite operates as a secondary wireless network. The objective is to maximize the sum rate of T-RIS-equipped LEO satellite communication using downlink non-orthogonal multiple access (NOMA) while ensuring the service quality of GEO cellular users. Our framework simultaneously optimizes the total transmit power of LEO, NOMA power allocation for LEO user equipment (LUE), and T-RIS phase shift design, subject to the service quality of LUE and interference temperature to the primary GEO network. To solve the non-convex sum rate maximization problem, we first adopt successive convex approximations to reduce the complexity of the formulated optimization. Then, we divide the problem into two parts, i.e., power allocation of LEO and phase shift design of T-RIS. The power allocation problem is solved using Karushâ Ȃ ŞKuhnâ Ȃ ŞTucker conditions, while the phase shift problem is handled by Taylor approximation and semidefinite programming. Numerical results are provided to validate the proposed optimization framework.

Effective detection of sophisticated cyber threats requires innovative frameworks capable of analyzing the complex and evolving behaviors of malicious software. The Temporal Behavior Chain Analysis (TBCA) framework introduces a novel methodology for identifying ransomware activity through the examination of temporally dependent system interactions. By constructing directed acyclic graphs to represent behavioral sequences, the framework captures subtle anomalies that traditional detection methods fail to identify. Machine learning techniques, including Hidden Markov Models and neural networks, enhance the framework's ability to discern malicious patterns from benign behaviors with high accuracy and efficiency. Evaluations demonstrated superior detection rates across multiple ransomware families, including LockBit, BlackCat, and Hive, even when advanced obfuscation techniques were employed. The modular design of TBCA ensures seamless integration into existing infrastructures while maintaining scalability for deployment in diverse operational environments. Resource efficiency was validated through extensive scalability testing, which confirmed the framework's capability to process high volumes of data with minimal computational overhead. Comparative analyses revealed significant improvements over traditional detection approaches, particularly in handling emerging and previously unseen ransomware variants. Temporal modeling of sequential dependencies enables early detection, reducing potential system damage and facilitating real-time mitigation strategies. The integration of adaptive decision-making mechanisms minimizes false positives, enhancing reliability in high-stakes environments. Cross-platform evaluations further highlighted the robustness of the framework, demonstrating its effectiveness on Windows, Linux, and macOS systems. The TBCA framework establishes a comprehensive foundation for advancing proactive cybersecurity defenses against increasingly complex and dynamic threats.

Parkinson's Disease (PD) is a progressive neurological disorder that affects both motor and nonmotor functions. A critical aspect of PD is postural instability, which is associated with a higher risk of falls and significant decline in mobility. The onset of postural instability is frequently identified through subjective clinical assessments such as the pull test, which aligns with the transition from Stage II to Stage III on the Hoehn and Yahr (H&Y) scale, a measure of general motor impairment. This study aims to pinpoint this transition by providing an objective quantification of the pull test, to prevent falls and manage severe motor symptoms via early detection. Sixty objective metrics were characterised from acceleration data using inertial measurement units (IMU) during pull tests conducted with 30 PD patients to differentiate between H&Y Stages II and III. Significant differences in feature distributions were identified, with two features achieving a ROC AUC of 0.95: chest to lumbar maximum acceleration ratio, and power ratio of 8-12 Hz versus total acceleration power. This paper demonstrates the ability of certain IMU-derived features to distinguish between H&Y Stages II and III, presenting more straightforward and objective measurements compared to conventional pull test metrics. Furthermore, the study explores the inconsistencies in the administration of the pull test by different examiners and shows that even with these inconsistencies, objective measures can still effectively differentiate between stages. To the best of our knowledge, this work represents the first extensive attempt to objectively quantify the pull test for this type of classification.

Contextual pathway reconfiguration introduces a dynamic approach to enhancing the performance and adaptability of transformer-based architectures in multimodal settings. By incorporating hierarchical gating mechanisms and auxiliary embedding layers, the proposed framework efficiently prioritizes and synthesizes information across diverse input modalities, addressing long-standing challenges in feature alignment and integration. Extensive experiments demonstrated consistent improvements in performance metrics across a variety of tasks, including image captioning and video question answering, showcasing the model's ability to generalize effectively while maintaining computational efficiency. The reinforcement learning-driven pathway evaluation mechanism emerged as a critical component, enabling real-time adjustments to computational pathways based on contextual input demands. Comparisons with traditional transformer methodologies highlighted the modular design's superiority in both scalability and robustness, particularly in handling noise-prone and unbalanced datasets. An in-depth analysis of latency and throughput further demonstrated the model's potential for deployment in high-demand, real-time applications. Error analysis revealed specific cases of improvement in cross-modal reasoning while identifying areas where input noise posed residual challenges. Ablation studies validated the significance of individual architectural components, confirming their contribution to overall model performance. The novel approach to dynamically reconfiguring attention mechanisms has broad implications for advancing multimodal applications, offering a versatile solution to integrate textual, visual, and auditory data. Additionally, energy efficiency assessments demonstrated the model's practicality for both training and inference scenarios, enhancing its viability for large-scale use. The findings present a compelling case for redefining multimodal transformer architectures to balance precision, efficiency, and adaptability, paving the way for more robust and versatile implementations.

Pulse-Width-Modulated (PWM) thermo-optic tuning in silicon photonics requires power supplies with low quiescent current, compact size, and fast PWM output voltage. However, the transient performance of conventional outputcapacitorless low-dropout regulators (OCL-LDOs) is limited by their loop bandwidth and slew rate, which prevents them from achieving fast PWM tuning. Inspired by application-specific designs (e.g., GPU, TPU, etc.) that sacrifice general performance for specific performance improvements, this paper presents a gate-switching (GS) LDO that is optimized for fast PWM tuning instead of general transient response. In the proposed design, the speed of PWM tuning is independent of the global feedback loop of the LDO and is only limited by the discharging speed of the pass transistor gate, which is accelerated through a synchronized gate discharging (SGD) block. This design has been implemented in a 65 nm standard CMOS process, occupying an active area of 2800 µm 2. With a unity-gain-frequency (UGF) of 27.8 MHz, this design achieves an average settling time of 15 ns when transitioning between zero load current and the maximum output current, significantly faster than its 1/UGF value. With a 1.2 V input voltage, it can generate a 1-MHz, 1-Vpp output voltage with a duty ratio ranging from 2.5% to 99.5% with a 50 Ω heater load.

Power delay profile (PDP) of a wireless channel has been widely studied using different approaches and measurements. In this paper we take a new approach by applying Friis transmission equation to derive the PDP in a Gaussian scattering scenario. Using this approach we notice that if the scattering spread is large or the distance between the base station and the mobile station is small, the PDP has a peak at a delay greater than the line of sight delay. The obtained PDP is able to explain the peak reported in measurements and is consistent with the delay probability density function. Further, the profile comes closer to the PDP reported in the literature when either the distance between the base station and mobile station is large or the scattering spread is small.

In this paper, we propose a novel Privacy-Preserving clearance mechanism for Local Energy Markets (PP-LEM), designed for computational efficiency and social welfare. PP-LEM incorporates a novel competitive game-theoretical clearance mechanism, modelled as a Stackelberg Game. Based on this mechanism, a privacy-preserving market model is developed using a partially homomorphic cryptosystem, allowing buyers’ reaction function calculations to be executed over encrypted data without exposing sensitive information of both buyers and sellers. The comprehensive performance evaluation demonstrates that PP-LEM is highly effective in delivering an incentive clearance mechanism with computational efficiency, enabling it to clear the market for 200 users within the order of seconds while concurrently protecting user privacy. Compared to the state of the art, PP-LEM achieves improved computational efficiency without compromising social welfare while still providing user privacy protection.

As the world searches for groundbreaking, unconventional computing technologies, especially for intelligent edge applications, biological AI is emerging as an energy-efficient, robust, and reliable alternative. Researchers have unveiled the immense computing capacity inherent in biocomputing elements such as bacterial cells. The computing power of bacteria can be harnessed through Gene Regulatory Neural Networks (GRNNs). Biofilms, acting as sophisticated collections of GRNNs, leverage the natural distributed computing architecture with capabilities like parallel processing and analog computing in individual cells while consuming very little energy relative to conventional computing systems. This study introduces the concept of Biofilm Living AI Devices (BLAIDs), which proposes engineering biofilms using optogenetics to function as self-sustaining AI edge devices that interface with modern telecommunications architectures. Our simulation-based analysis demonstrates the computing complexity and reliability of BLAID, establishing it as a compelling candidate for the next generation of low-energy computing and advanced AI technologies.

A framework combining data driven deep learning with domain-driven controller logic is introduced in this paper. This is done by encapsulating the controller logic into a digital twin that generates Signal Phasing and Timing (SPaT) information created by the controller given the controller logic and actuation data. The digital twin is designed as an extensible object-oriented program that can run provided the controller data. The controller data are used in two ways. First, it is used to initialize the digital twin by extracting the controller settings from the data either through explicitly mentioned variables in the data or implicit variables inferred from the data. Second, a deep learning multilayer perceptron (MLP) model is used with the data to estimate the actuations. Actuations are then fed to the controller digital twin for the digital twin to provide SPaT predictions. This approach is tested on a single intersection and provides a much better overall prediction performance than other approaches with perfect predictions 80 percent of the time and within 1 second of actual switch times 85 percent of the time. The mean absolute error and mean absolute percentage error in this approach are 0.67s and 7.61 percent respectively, resulting in a more than 58 percent reduction in the prediction error compared to other models presented in the literature.

Semantic and goal-oriented (SGO) communication is an emerging technology that only transmits significant information for a given task. Semantic communication encounters many challenges, such as computational complexity at end users, availability of data, and privacy-preserving. This work presents a TinyML-based semantic communication framework for few-shot wireless image classification that integrates split-learning and meta-learning. We exploit split-learning to limit the computations performed by the end-users while ensuring privacy-preserving. In addition, meta-learning overcomes data availability concerns and speeds up training by utilizing similarly trained tasks. The proposed algorithm is tested using a data set of images of hand-written letters. In addition, we present an uncertainty analysis of the predictions using conformal prediction (CP) techniques. Simulation results show that the proposed Semantic-MSL outperforms conventional schemes by achieving 20% gain on classification accuracy using fewer data points, yet less training energy consumption.

The long-term drift is well known as one of the most challenging issues regarding gas sensing and applications involving electronic noses, undermining the reliability and accuracy of these systems over time, as the pattern recognition models lose their capability of performing gas discrimination. In this work, we adopt a probabilistic view of the drift problem. This perspective yields a novel univariate scheme for drift compensation, which is grounded on probabilistic modelling. The potential usefulness of the proposed approach is highlighted through experimental results on the gas classification task, which reveal a superior and more stable performance when compared to traditional component correction multivariate methods. Besides its stable performance, the proposed method also presents several desired attributes, such as computational simplicity, continuous adaptability and suitability for online applications. The experiments were conducted using a dataset lasting 7 months, comprising measurements from three different gases at different concentration levels in an array containing 17 polymeric sensors.

Multi-jet impingement-based cooling manifolds, functioning as integrated networks for liquid delivery and effusion, have become increasingly important due to the rising thermal management demands of high-performance computing chips. As these chips generate significant thermal load during operation, efficient cooling solutions are essential to maintain performance and prevent thermal-related reliability issues. This study focuses on the conceptual design and assessment of an innovative manifold, with an emphasis on its thermal and hydraulic performance. A 3D-printed serpentine manifold with distributed inlets and outlets is considered, where hot and cold liquids are separated by partition walls, resulting in a compact and lid-compatible design. The manifold is designed to cool a heated area of 10 mm × 10 mm, with the heat flux reaching up to approximately 415 W/cm². To assess the performance, steady-state temperature and pressure drop across the manifold are experimentally studied under different coolant flow rates and heat fluxes. The thermo-hydrodynamic performance of the cooler is compared with existing similar designs in the literature, focusing on single-phase coolant operation. Additionally, the study extends the analysis to include two-phase operation, exploring the effectiveness of the manifold. This innovative, compact manifold design has the great potential to act as an integrated heat spreader, offering a promising cooling solution for high heat flux applications while maintaining heat transfer efficiency.

A compact spiral cavity backed substrate integrated waveguide (SIW) Multiple Input multiple output (MIMO) antenna is presented in this paper. The edge-shaped spiral on top of the SIW cavity acts as a dipole antenna. The dual spiral arms are excited from their symmetrical connecting center. The unit cell comprising of dual spiral arms is rotated such that each unit cell is orthogonal to each other forming a compact 4x4 MIMO SIW antenna. The proposed design shows a wide bandwidth of 930 MHz (13.74 GHz to 14.67 GHz) and 67.68% impedance bandwidth. The overall size of proposed MIMO SIW antenna is 0.9 λ o x 0.9 λ o x 0.024 λ o , where λ o is the operating wavelength. A good return loss of-18.4 dB at 14.17 GHz is achieved. The metal pins of plus shape at the center improves the isolation to 19.6 dB at resonant frequency. A pattern diversity is achieved by the top spiral radiating arms and it's complementary structure at the bottom of the cavity. The beam-width of the proposed antenna is 90° varying from-45 deg to +45 deg, useful for reliable signal transmission and reception. Thus proposed antenna design is a symmetrical compact design working at Ku band suitable for radio telescope application.

The problem of robust quickest change detection (QCD) in non-stationary processes under a multi-stream setting is studied. In classical QCD theory, optimal solutions are developed to detect a sudden change in the distribution of stationary data. Most studies have focused on single-stream data. In non-stationary processes, the data distribution both before and after change varies with time and is not precisely known. The multi-dimension data even complicates such issues. It is shown that if the non-stationary family for each dimension or stream has a least favorable law (LFL) or distribution in a well-defined sense, then the algorithm designed using the LFLs is robust optimal. The notion of LFL defined in this work differs from the classical definitions due to the dependence of the post-change model on the change point. Examples of multi-stream non-stationary processes encountered in public health monitoring and aviation applications are provided. Our robust algorithm is applied to simulated and real data to show its effectiveness.

This paper presents a novel framework for real-time object detection in Wireless Mesh Networks (WMNs) using AI driven drones. The system integrates YOLOv5 for object detection, a hybrid greedy and reinforcement learning-based routing algorithm for efficient data communication, and a scalable multihop mesh topology. MATLAB-based simulations evaluate energy consumption, detection accuracy, and network connectivity, showcasing significant performance improvements. The results demonstrate the system's ability to detect objects accurately, maintain robust connectivity, and optimize energy consumption, even in dynamic environments. Future work includes expanding the system to integrate 5G and satellite networks and conducting field tests.

CHARGE (Charged Hyperspherical Autoencoder with Repulsive Gradient Encoding) is a novel autoencoder designed with a unique latent space that is hyperspherical and of n dimensions. The model leverages a specialized loss function based on Coulomb's Law or the Inverse Square Law, promoting a more effective and efficient latent space. This approach enables CHARGE to perform better on generative AI tasks compared to traditional Variational Autoencoders (VAEs), as it is deterministic rather than probabilistic. The results show that CHARGE provides superior performance in generating realistic data and encoding meaningful latent representations, making it a promising candidate for various generative tasks in AI.