The paper introduces the Neural Mesh, a novel architecture that serves as an alternative to conventional neural network models widely used in contemporary AI systems. In parallel, the paper introduces the term Logical Synthesis Model (LSM) as a complement to Large Language Models (LLMs). While LLMs primarily rely on statistical representations and language-based reasoning to acquire and generate knowledge, Neural Mesh-based LSMs focus on structured representations and engineering reasoning to represent, analyze, synthesize, and control systems. LLMs excel at learning and reasoning over human knowledge expressed through language, whereas LSMs aim to learn structured representations that support the “understanding” of physical systems. In this sense, LLMs and LSMs may play roles analogous to the cerebrum and the cerebellum in biological intelligence. The novelty of the paper lies in bridging and combining the concepts of TP models, TP model transformation, and neural-network architectures. The proposed Neural Mesh is mathematically equivalent to the well-established TP model function family. The TP model transfor-mation provides a framework for constructing TP models with unique features that are particularly advantageous for systems and control applications. The contribution of the Neural Mesh is that it represents these unique features in the form of a special neural-network architecture, where these features are expressed as trainable neural connections and parameters. As a result, elements that are traditionally determined through an offline TP model transformation become directly tunable through learn-ing, while preserving the mathematical structure and control-theoretic advantages of the TP model framework. Therefore, the Neural Mesh representation opens a future research direction in which the TP model transformation is effectively embedded into the training process itself. Rather than generating a TP model through a subsequent offline transformation, the corresponding TP-model components are directly constructed and tuned during system identification via neural-network training tools.
Polarization manipulation in visible wavelengths has an important role in many applications, such as free-space quantum communication and quantum cryptography. In this study, a polarization rotator based on a tilted slot waveguide by 45-degree is proposed and designed. Its principle is to rotate a +45°/-45° input polarized light to TM0/TE0 output polarized light. The simulation results show an extinction ratio of 45.23 (41.17) dB and insertion loss below 0.62 dB at 700 nm wavelength. The bandwidth for an extinction ratio higher than 20 dB is 40 nm. In addition, the device tolerance is analyzed.
Emerging technologies offer validation and authentication solutions in the field of audiovisual content creation. Visible or invisible watermarking, embedded metadata, and digital signatures can be used to maintain the validity and creditability of still images and video data. The Coalition for Content Provenance and Authenticity (C2PA) was established to create an open source framework and to provide technical solutions for image capture, processing, delivery, and verification. The leading market players in hardware and software development set the goal of applying encrypted metadata information to guarantee the authenticity of the data. Currently, only a few devices and applications are available and have been implemented based on this technology. This paper gives an introductory overview of the recent state, highlighting advantages, drawbacks, available implementations, and future perspectives on research directions.
Fuzzy Linguistic Signatures (FLS) extend the concept of Fuzzy Signatures (FSigs) by introducing linguistic variables as qualitative descriptors within a hierarchical fuzzy structure. Although fuzzy signatures have been successfully applied in various domains, their reliance on numerical membership degrees limits their ability to model subjective or linguistically defined information. This paper establishes a formal mathematical frame-work for FLS by defining a family of fuzzy linguistic signatures equipped with suitable linguistic aggregation operators and a partial ordering relation among linguistic values. Furthermore, meet-and-join operators are introduced to demonstrate that FLS satisfies the properties of a lattice as an algebraic structure. Consequently, fuzzy linguistic signatures provide an expressive representational framework capable of handling qualitative, human-like reasoning.
The deployment of Encrypted Client Hello (ECH) challenges TLS fingerprinting, a widely used approach for encrypted malware detection, by encrypting the handshake fields these methods rely on. This paper presents a systematic evaluation of flow-based statistical features as a handshakeindependent alternative to fingerprinting. Through validation against the official JA4+ implementation, we establish limitations in fingerprinting approaches for this corpus: only 64.9% of malware families possess unique signatures, placing an inherent ceiling on achievable recall in our evaluation. We evaluate flow-level features—packet counts, timing patterns, and size distributions—across 27 experimental configurations on a dataset of 16,542 flows spanning 101 families (59 malware and 42 benign applications). Random Forest classifiers using combined flow statistics and sequential packet length features achieve 98.11% F1-score for binary malware detection with 97.22% recall, substantially exceeding fingerprinting’s theoretical recall bound of 64.9%. For fine-grained family identification, we obtain 54.81% macro F1 across 101 classes and 48.71% macro F1 for malwareonly attribution, demonstrating that flow-based methods retain meaningful discriminative power where fingerprinting abstains. Across all tasks, Random Forest consistently outperforms neural networks and k-NN, with performance gaps widening in complex multiclass scenarios. These findings highlight flow-based classification as a practical and reproducible approach that can help maintain network security visibility as ECH deployment progresses, showing that behavioral traffic patterns are expected to provide durable signals for detection even as handshake fields become encrypted
Software defect prediction plays a crucial role in ensuring the quality and reliability of software systems. Feature selection, the process of identifying the most relevant features from a large set of potential features which is essential for building effective defect prediction models. In this paper, we propose a novel feature selection model based on the RelayPursuit- Vathana (RP-Vathana) optimization algorithm, inspired by relay races and pursuit dynamics in biological systems. The proposed model aims to identify an optimal subset of features for software defect prediction, maximizing the predictive performance of the resulting classification model. The RP-Vathana algorithm was integrated with a Naïve Bayes classifier and benchmarked on three datasets (PC5, JM1, KC2) to validate its effectiveness in feature selection for defect prediction. The results show that RP-Vathana significantly outperforms existing wrapper-based methods, obtaining mean accuracies of 94.28%, 93.69%, and 96.35% on PC5, JM1, and KC2, respectively, compared to the 83−90% range of rival techniques. While the parameter-free design improves usability, the algorithm's performance on highly noisy or very small datasets warrants future investigation into hybrid extensions for enhanced robustness.
The transition from 4G to 5G networks was necessitated by fundamental limitations in spectral efficiency and data capacity inherent to the existing framework. Fifthgeneration (5G) systems address these constraints by capitalizing on key enabling technologies, such as mmWave spectrum, multiple-input multiple- output (MIMO), massive MIMO (mMIMO), beamforming (BF) or Precoding. This paper investigates a multi-layer transmission scheme employing MIMO with Precoding (SVD) as a cooperative technique to enhance downlink (DL) data transmission performance in an enhanced mobile broadband (eMBB) scenario. The study operates within the standard 5G mmWave frequency band (FR2 at 40 GHz). We differentiate between key performance metrics: the user-experienced data rate (or throughput) Measured at the Receiver (Rx) and the peak theoretical data rate (or Bit Rate) Measured at the Transmitter (Tx). Simulation results, conducted using MATLAB, demonstrate that the proposed approach significantly improves both the achievable throughput and spectral efficiency within a fixed bandwidth. Throughput is evaluated in absolute terms (Mbps) and as a normalized percentage of the peak theoretical data rate (Bit Rate). The core of this study examines the impact of the number of spatial data streamng layers on a 5G-NR system performance. While increasing the transmission layers enhances the potential peak data rate at the transmitter, it concurrently elevates the bit error rate (BER) at the receiver, ultimately degrading the net throughput. This underscores the necessity for advanced receiver-side technologies, such as MIMO processing, to counteract high-path loss and other impairments prevalent at mmWave frequencies. The results confirm that augmenting the number of antennas in the MIMO configuration effectively mitigates this limitation. It improves the overall throughput and reduces the received BER by enhancing spatial diversity and signal recovery capabilities.
The exponential growth of Internet of Things (IoT) devices and the growing demand for resource-intensive applications have introduced significant challenges in computation, storage, and network efficiency. Although cloud computing provides partial relief, its centralized nature leads to unacceptable latency for delay-sensitive applications. Multi-access Edge Computing (MEC), especially with the advent of 5G, has emerged as a compelling solution by relocating computation closer to data sources, thereby reducing latency and improving responsiveness in applications such as smart agriculture, autonomous vehicles, augmented reality, and telemedicine. However, efficient workload offloading in MEC environments remains complex due to system heterogeneity, varying application requirements, and limited edge resources. This paper proposes a novel neural network-based approach to computation offloading in MEC, integrating workload allocation and resource management while accounting for application delay sensitivity, processing capacity, and communication constraints. The proposed model enables driving offloading decisions, adapting to fluctuating system states without relying on complex mathematical formulations. Simulation results demonstrate that the approach significantly reduces service time and enhances resource utilization, ensuring responsiveness for modern IoT applications. This research underscores MEC’s potential to meet the rising computational and latency demands of next-generation IoT infrastructure.
Short-packet transmission is becoming crucial for satellite services that cannot rely on long codewords to hit the required latency and reliability. This study investigates ratesplitting multiple access (RSMA) in that context and builds a finite-blocklength (FBL) model for a downlink satellite-terrestrial link affected by Shadowed-Rician fading. We obtain closed-form approximations for the block error rate (BLER) of both the common and private streams, explicitly incorporating imperfect successive interference cancellation (ipSIC) at the receivers. Compared with power-domain non-orthogonal multiple access (NOMA), RSMA exhibits more stable BLER the common stream helps dampen residual interference due to ipSIC and the short-packet effect–so RSMA generally needs less transmit power to attain the same error targets. Numerical results validate the analysis and demonstrate consistent RSMA advantages across a wide range of transmit powers, blocklengths, shadowing severities, and antenna configurations. The results suggest that RSMA is a very promising option for future satellite systems that need to provide reliable, low-latency, and short-packet communications in view of realistic SIC imperfections.
The Improved Rapidly-exploring Random Tree with Reduced Random Map Size (IRRT*-RRMS) algorithm was previ- ously developed to find collision-free paths for mobile robot path planning. Given the excellent performance of Transformer Neural Networks with sequential data, we propose an encoder-decoder transformer model combining a Vision Transformer (ViT) as the encoder and a time-series forecasting module as the decoder to learn the path planning algorithm. The novelty of this paper lies in developing a model supervised by a dataset generated from the IRRT*-RRMS algorithm and using this trained model for the path planning task. The trained model efficiently predicts intermediate points between the desired starting and goal points. The performance was validated on a real robot, demonstrating that the trained model required less computation time compared to the IRRT*-RRMS algorithm.
In this paper, we propose quantum orthogonal frequency division multiple access (Q-OFDMA), a novel quantum communication scheme designed to overcome the fidelity limitations imposed by noise in multi-user quantum networks. Inspired by its classical counterpart, Q-OFDMA employs the quantum Fourier transform (QFT) and its inverse (IQFT) to encode and decode information across quantum channels. We evaluate our model under both a depolarization channel and a generalized noise model that interpolates between depolarizing and phase-damping noises. The simulation results conducted on Qiskit platform demonstrate that Q-OFDMA outperforms the reference model, achieving superior average fidelity across varying qubit counts and noise levels.
Recent advances in large language models (LLMs) have enabled highly human-like text generation, raising concerns related to misinformation, authorship verification, and academic integrity. Current approaches for detecting LLMgenerated text suffer from several limitations, including limited robustness to linguistic diversity, sensitivity to text length variations and paraphrasing, weak domain generalization, and high computational cost. To address these challenges, this paper proposes a hybrid framework for detecting LLM-generated text that integrates syntactic and statistical features with deep semantic representations learned using GloVe embeddings, Convolutional Neural Networks (CNNs), and Bidirectional Long Short-Term Memory (BiLSTM) networks. By combining linguistic cues with contextual semantics, the proposed model captures both structural and semantic patterns to distinguish human-written text from LLM-generated content. Experiments conducted on the ChatGPT Research Abstracts and ElectAI datasets demonstrate strong cross-domain generalization and robustness to text length variations and paraphrasing. The proposed framework achieves an accuracy of 98.63%, an F1-score of 98.66%, and a minimum false positive rate (FPR) of 0.01. These results indicate the effectiveness, stability, and reliability of the framework for detecting LLM- generated text.
The Internet of Things (IoT) has transformed device connectivity with the smooth interfacing for realtime data exchange across multiple applications, from smart homes to industrial automation. Nonetheless, as networks under IoT, especially those using the routing protocol for low-power and lossy networks (RPL), continue in their expansion, the security penetration becomes much more evident. One of the major security constraints is suboptimization attacks-they negatively affect network performance, scalability, and data integrity. These attacks impede the very efficiency of the IoT systems, thereby making it so challenging for the systems to be secured and maintained successfully. Traditional IDS and cryptographic solutions are seldom fit-for-purpose in dynamic IoT environments, which opens up the need for the ability to provide scalable and energy-aware security solutions. This review investigates and surveys existing IDS, cryptographic solutions, and machine learning techniques targeting and working against such threats. It puts forth an integrated solution where an adaptive IDS is combined with scalable, energy-efficient, real-time anomaly detection to make IoT networks more resilient to sub-optimization attacks. According to this study, dynamic, contextaware safety measures are essential, as they are capeble of adressing the new challenges arising from IoT environments.
High-accuracy automatic modulation classification (AMC) is essential for spectrum monitoring and interferenceaware access in future 6G systems [1]. We propose AMCTransformer, which tokenizes raw I/Q sequences into fixedlength patches, augments them with learnable positional embeddings, and applies multi-layer, multi-head self-attention to capture global temporal–spatial correlations without handcrafted features or convolutions. On RadioML2018.01A, our model achieves 98.8% accuracy in the high-SNR regime (SNR at least 10 dB), showing higher accuracy than a CNN and a ResNet reimplementation by 4.44% and 1.96% in relative terms; averaged across all SNRs, it also improves upon MCformer, CNN, and ResNet baselines. Consistent gains are observed on the RadioML2016.10A dataset, further validating robustness across benchmarks. Ablations on depth, patch size, and head count provide practical guidance under different SNR regimes and compute budgets. These results demonstrate the promise of transformer-based AMC for robust recognition in complex wireless environments.
Quantum communications promises major changes in today’s communication networks by sending qubits over long distances. These qubits enable large-scale quantum computing or information-theoretically secure distribution of symmetrical keys. One of the main enablers is quantum teleportation, which makes sending quantum information between two nodes possible even when they are far apart. From these nodes, one can build a larger quantum network, but due to the nature of quantum physics, certain tasks that are well understood in classical networks, such as routing, cannot be handled in a similar way. Our work focuses on modifying a previously created model for a ring-like quantum network and assessing the effect of introducing a new node type. Our results show that this node can alter properties of the underlying network. We also look at the possibilities of modeling the capacity of the network as well as the availability of the newly introduced edges, which open interesting questions for future research.
Orthogonal Time Sequency Multiplexing (OTSM) represents a pivotal advancement in wireless communication technology. Nevertheless, its high Peak-to-Average Power Ratio (PAPR) imposes significant constraints on its practical applications and future development. The definition of PAPR refers to the ratio of the maximum instantaneous power to the average power of a signal, and it is commonly used to assess the performance of high-power amplifiers. When PAPR values are excessively high, they reduce the efficiency of high-power amplifiers and increase the complexity of the transmission system. To mitigate this challenge, this paper explores and evaluates the efficacy of the Selective Mapping (SLM) technique for enhancing PAPR performance in OTSM systems. Leveraging the unique two-dimensional data structure inherent to OTSM, a specialized SLM approach is introduced in this paper. The proposed SLM method incorporates a Phase Generation Mechanism (PGM) that utilizes a pre-constructed perturbation phase matrix. This matrix undergoes cyclic shifts to produce multiple perturbation phase matrices. To assess the effectiveness of the proposed SLM technique, this paper investigates three distinct perturbation phase matrix generation mechanisms: Zadoff-Chu Transform (ZCT) matrices, Discrete Cosine Transform (DCT) matrices, and Randomly Generated Phase (RGP) matrices. Additionally, for evaluating PAPR performance improvement, the Complementary Cumulative Distribution Function (CCDF) is used, a statistical method that estimates the probability of high PAPR occurrences. Simulation results indicate that the RGP-based phase generation mechanism consistently outperforms the other methods in achieving significant PAPR reduction.
Intellectual disability (ID) involves deficits in intellectual and adaptive functioning specifically related to conceptual, social, and practical life domains. Computer-assisted training programs have been shown to enhance adaptive competencies of adolescents with ID, thus helping them to better manage their everyday life and to foster their social inclusion and integration. The present work is aimed at reviewing the existing literature on computer-assisted interventions devoted to adolescents with ID highlighting their actual efficacy and strengths. Implications for future research and practice are discussed.
This paper presents a systematic methodology for evaluating the quality of technical reviews in software development, aiming to address the issue of ineffective reviews and their impact on product quality. The methodology, grounded in literature review, identifies key factors for highquality reviews, including planning, reviewer commitment, and meaningful feedback. It emphasizes linguistic clarity and accuracy, drawing lessons from the Therac-25 case, where poor documentation contributed to fatal accidents. The paper analyzes specific linguistic issues like negative structures, passive voice, and terminology, highlighting their impact on comprehension. The methodology's effectiveness is validated through a Lean Six Sigma project in a large software development company, resulting in significant improvements. These include a 60% improvement in the Technical Review Quality KPI, a 29% reduction in customer-reported faults related to reviews, and the elimination of TL9000 non-conformities. This case study demonstrates the practical applicability of the framework and its potential for significant impact. The paper concludes by highlighting the importance of linguistic considerations in ensuring safer and more effective software products. Future research directions include extending the methodology to other types of artifacts and exploring textual analysis of reviewer feedback for deeper insights into the review process.
Understanding spatial perception is crucial in virtual environments since it influences navigation abilities. To better understand how human characteristics combined with immersion levels influence exocentric distance estimation, this study was conducted. We have implemented a virtual environment for the desktop display and the Gear VR head-mounted display, in which we assessed these skills of 229 university students. 157 used the former, while 72 used the latter display device. The results show that human characteristics combined with the two display devices as well as a virtual scale have significant effects on exocentric distance estimation. The findings can help the development of more accessible virtual environments in the future.
This paper presents novel signal processing methods to enhange the reception performance of Gaussian Minimum Shift Keying (GMSK) signals from pico- and nanosatellites, emphasiz-ing software-defined approaches over hardware upgrades. Atypi-cal filtering techniques, including phase-domain median and FIR filtering, as well as polarization-diverse multi-channel methods, are explored and evaluated. Real-world experiments were con-ducted uling coherently sampled dual-polarization channels from a ground station in Szeged, Hungary, receiving transmissions from the MRG-100 satellite. Various single- and multi-channel preproc-essing strategies were benchmarked asing packet decoding success and bit-error rates. Results show that non-linear phase filtering and blind source separation techniques, #otably FastICA, significantly increase the number of correctly decaded packets - achieving up to a 16% infprovement compared to conventional demodulation with-out preplocessing. This study demonstrates the utility and relative independence of these methods and highlights their potential for improving satellite data throughput with no hardware modifica-tion. These techniques are suitable for integration into existing ground stations to enhance data reception performance.
