To establish stable vertical communication links for Non-Terrestrial Link (NTN), such as between ground stations and satellites, it is essential to mitigate the impact of weather conditions—such as clouds and rainfall—on signal propagation in NTN links. As a countermeasure in satellite optical communications, site diversity techniques are employed to avoid the effects of cloud coverage. This paper presents the results of an investigation into the impact of weather on optical communication, based on observational data collected at ground level and meteorological data obtained from satellites.
Electric bicycles (e-bikes) have gained significant popularity due to their eco-friendliness, affordability, and accessibility. Moreover, e-bikes serve as an inclusive transportation solution for individuals with limited physical stamina or mobility constraints, enabling longer and more frequent travel than traditional bicycles.This paper introduces an Advanced Rider Assistance System (ARAS) designed to enhance cyclist safety on smart e-bikes. The system is implemented on a prototype smart e-bike platform developed at the German University in Cairo (GUC), equipped with a resource-constrained Raspberry Pi 4. It leverages real-time embedded vision to detect road hazards.The proposed ARAS builds upon previous work in pothole detection, speed bump recognition (both marked and unmarked), cyclist alerting mechanisms, and traffic sign interpretation. A custom dataset was collected to train and evaluate several state-of-the-art object detection architectures, including YOLOv11, YOLOv12, RT-DETRv2, and D-FINE. Our findings show that the transformer-based D-FINE-N model provides the optimal performance-speed trade-off, demonstrating a superior balance of precision (88.7%) and recall (87.5%) and a leading F1-score of 88.1% and mAP@50-95 of 49.3%. Moreover, after optimization and conversion to the ONNX format, D-FINE-N achieved the highest inference speed, reaching 2.75 FPS on the target hardware. This research demonstrates the viability of deploying modern deep learning models for ARAS applications on low-power edge devices and highlights the effectiveness of D-FINE-N for real-time safety systems.
The electromagnetic effective degrees of freedom (EM EDOF) provide an insightful measure for characterizing the performance of multiple-input-multiple-output (MIMO) wireless communication systems. The dyadic Green’s function (DGF) is widely used as a benchmark tool for modeling ideal EM propagation due to its analytically tractable closed-form solution. This paper proposes a scatterer-parameterized EM channel model based on the DGF to characterize the EM EDOF in MIMO systems. The proposed EM model is first validated by comparison with results obtained using the method of moments (MoM). Then, the impact of distinct polarizations on the EM EDOF is investigated, highlighting the advantage of full polarization over transverse polarization in a scattering environment. Finally, the EM EDOF is analyzed through a sensitivity analysis of EM model parameters and further evaluated under varying scatterer-antenna distances. Numerical results show that near-field spatial modulation induced by scatterers increases the EM EDOF, with the outer scatterer regions contributing more prominently, while the effect weakens and converges to the free-space value in the far field. The proposed model offers an efficient tool for analyzing and optimizing MIMO capacity and spatial DOF.
Visual odometry is a crucial technology in mobile robotics, enabling the estimation of motion through continuous image observations. Traditional algorithms use geometric elements like point and line features to establish correspondences between frames for real-time pose estimation. While point features are simple but susceptible to external factors, line features offer stability but suffer from extraction speed and distribution issues. Effectively merge line feature into odometry becomes a highly promising direction for improving accuracy. To this end, this paper proposes HS-VO, a homogenized structural visual odometry. It incorporates quick line extraction and stable homogenization schemes, introducing Manhattan Axes (MA) for scene geometric constraints. Additionally, to effectively extract MA, we design a Breadth First Search (BFS)-based method for plane normal vector extraction. To verify the performance of HS-VO, we conduct experiments on public datasets. The results show that HS-VO significantly improves both speed and accuracy compared to other approaches.
Radar communication integration based on OFDM (Orthogonal Frequency Division Multiplexing) waveform is a promising approach to achieve dual functions of broadband wireless communications and radar sensing. DFTs-OFDM (Discrete Fourier Transform spreading-OFDM) radar using Zadoff-Chu (ZC) sequence is proposed because of ideal autocorrelation of ZC sequence and low PAPR (Peak to Average Power Ratio). However, DFTs-OFDM radar using ZC sequence has drawbacks of large range sidelobe which degrades multiple target detection performance and signal processing complexity in conventional time-domain based correlation detection. To solve these problems, this paper proposes novel frequency domain signal processing for DFT-OFDM radar using ZC sequence. It describes principle of the proposed frequency domain signal processing for DFTs-OFDM radar. It also evaluates range profile of DFTs-OFDM radar for various design parameters by computer simulations and demonstrates that the range sidelobe can be reduced to less than -60 dB and signal processing complexity is also reduced to less than 1/300.
6G Networks is scheduled to have its first release in 2030, as the first human-centric cognitive wireless network, supporting the United Nations Sustainable Development Goals (SDGs) and the Japanese concept entitled Society 5.0. Promoting digital inclusion, closing the usage gap, and ensuring trustworthy, explainable artificial intelligence (AI) and human-centric services will be key to 6G achieving social and economic progress. Thus, population awareness, transparency, and participation are paramount as AI and emerging technologies evolve. History shows how 5G disinformation fueled COVID-19 fears, revealing the deep link between technology, culture and society. This paper proposes a three-part framework for 6G market readiness based on machine AI integration encompassing cognitive and autonomous systems, a human-centric user-experience (UX) model, and market-driven core principles. The proposed framework offers guidance for engineering 6G products and use cases that align ethics with innovation, promote sustainable deployment, and support the Communication, Navigation, Sensing, and Services (CONASENSE) platform.
Interference management of 5G base stations (BSs) is indispensable for spectrum sharing with other communication systems, e.g., fixed satellite service. Notably, massive multi-input multi-output (mMIMO) systems form high-gain beams, causing larger interference with other systems. This study focuses on the interference mitigation for codebook-based 5G mMIMO systems and proposes a novel nullforming approach based on a codebook subset restriction (CBSR) mechanism. A BS determines a subset of 2D discrete Fourier transform (DFT) beams from the Type-I codebook defined in 5G, where the BS exploits the orthogonality of the DFT beams to form a horizontal or vertical null in a target direction. Then, the use of unselected beams is restricted. Results show that the proposed scheme can successfully form horizontal or vertical nulls only by using the CBSR framework in 5G. The interference reduction performance achieved using the proposed horizontal and vertical nullforming is over 16.7 dB and 14.1 dB, respectively, regardless of the target direction.
Multi-view 3D human pose estimation is a fundamental problem in computer vision with applications in human-computer interaction, AR/VR, robotics, and tactile internet, where accurate body motion capture is crucial for immersive and real-time interaction. Traditional approaches often rely on camera calibration at inference, typically through triangulation of 2D detections, which limits deployment in dynamic or unstructured environments. In this work, we propose a novel framework that leverages intermediate features from transformer backbones for multi-level, cross-view feature fusion. A teacher-student supervision strategy exploits camera extrinsics only during training to guide the learning of geometry-aware correspondences. Evaluations on the CMU Panoptic dataset demonstrate that the proposed method achieves state-of-the-art performance among calibration-free approaches.
The application of unmanned aerial vehicles (UAVs) in improving the coverage of terrestrial networks have been established as an important field in wireless communications. The realization of these systems is dependent on the positioning of the aerial base stations (ABSs) due to their high mobility in three dimensions (3D), which is complicated by the random measurement noise at the receivers. This paper estimates the ABS’s position using the golden section (GS) method and compares the throughput gain to terrestrial users served by the ABS, with baseline methods for the same system model. The ABS is associated with users with low downlink throughput from the stationary access point (AP), which is used by default. Additionally, the relationship between the throughput and user density is explored. System throughput of up to 110 Mbps and an average improvement to users’ data rate of 125%, are achieved.
This paper proposes a smart urban water management system using low-cost Internet of Things (IoT) hardware and an easy-to-use, robust Android application. Urban water supply systems face significant issues like water wastage and inefficiencies due to overuse in households and manipulation with analog water meters, resulting in supply shortages and imbalances. The notable features of the proposed system are cloud-based storage of individual usage data, remote supply control, and digital bill generation in a single application. The proposed system offers a fair billing system for urban water meters, reducing overhead and scams, promoting water conservation awareness, and improving efficiency and transparency in municipal water management. This smart solution promises a more balanced and fair water supply, paving the way for efficient water management in today’s smart cities.
This paper describes the role of Geostationary Earth Orbit (GEO) satellites in the 6G era, technological challenges, and research and development activity by using the Japanese Engineering Test Satellite 9 (ETS-9). In the 6G era, Non-Terrestrial Networks (NTN) will be inter-connected as Multi-Orbit network, as well as Terrestrial Network (TN) and NTN will be interconnected. Among them, GEO satellite has an important role as a communication hub in Multi-Orbit network. Core technologies are Software Defined Satellite (SDS) technology, optical high-speed communication, and network management. National Institute of Communications and Information Technology (NICT) has been conducting research and development of such core technologies, and will demonstrate them by using the Japanese Engineering Test Satellite ETS-9. The ETS-9 will also be used as in-orbit testbed hub for innovation and demonstration of various use cases for Beyond 5G/6G.
This paper evaluates three open-source tools—Batfish, pyATS, and SuzieQ—for automated network testing in programmable infrastructures. It compares their capabilities across different phases of the network lifecycle, highlighting use cases, integration methods, and limitations.
Accurate estimation of Liquid Water Content (LWC) is vital for understanding cloud microphysics and improving weather prediction. Our previous study [1] used a Bayesian Neural Network (BNN) with dual-frequency radar (35 GHz, 95 GHz) to estimate LWC, achieving high accuracy but being limited by a small dataset. In this study, we constructed a large-scale dataset of 129,357 samples using observations from the U.S. Department of Energy's Atmospheric Radiation Measurement (ARM) program at the Cape Cod site (2016-2017). After preprocessing-including temporal synchronization, altitude interpolation, noise reduction, and cloud mask extraction-four models were evaluated: Random Forest (RF), XGBoost (XGB), Decision Tree (DT), and Linear Regression (LR). RF and XGB achieved R-2 = 0.78 and 0.76, comparable to the previous BNN results, despite not relying on probabilistic inference. These findings demonstrate that a data-centric approach, leveraging large-scale ARM observations and ensemble learning, enables robust and practical LWC estimation without the computational cost of Bayesian models.
Orthogonal Time-Frequency-Space (OTFS) modulation offers robustness against Doppler frequency shifts in high-mobility scenarios. However, under fractional Doppler channels, its delay-Doppler domain channel matrix suffers from a significant increase in the number of equivalent taps. This leads to unacceptably high complexity for the traditional message passing (MP) detection algorithm. To address this problem, this paper proposes an improved signal detection algorithm. The algorithm first simplifies and reconstructs the signal model, then incorporates threshold screening and equivalent noise variance estimation. This approach significantly reduces the algorithm complexity while maintaining detection performance.
Device-to-Device (D2D) communication is a key feature of next-generation communication networks, enabling direct and reliable data exchange between devices in close proximity without routing through a base station. This reduces the traffic load on base stations and enhances overall network spectral efficiency. In this work, we address the challenges posed by the low-rank sparse nature of the D2D cascaded channel in a reconfigurable intelligent surface (RIS)-aided system. We propose a Sparse Bayesian Learning algorithm for efficient sparse channel estimation and employ a gradient descent method to optimize the phase shifts of the RIS elements. This joint approach to sparse channel estimation and passive beamforming aims to maximize the achievable sum rate of the RIS-aided D2D system. Simulation results demonstrate that the proposed algorithms achieve significant performance improvements in both channel estimation accuracy and spectral efficiency compared to existing methods.
The safety and operational efficiency of railway infrastructural heavily relies on the condition of railway wheelsets and effective defect detection. Traditional inspection methods such as visual inspection and ultrasonic testing show limitations in terms of precision, operational efficiency and scalability. In this study, a novel approach is presented for real-time wheelset defect detection using state-of-the-art deep learning model: Neural Compact Refined Architecture-You Look Only Once model (NCRA-YOLO). A large dataset consists of 7487 high-resolution annotated images of different wheelset. The dataset used for training and testing utilized advanced data augmentation techniques to enhance performance. The model demonstrated superior accuracy of 100% in fracture and 83% in spot class. This model is the most effective solution with only 1.3 million parameters and 3.8 GFLOPS that proves its effectiveness in real time defect detection. The model achieved a precision of 94.7% and a mean average precision at IoU thresholds of 50% to 95% (mAP50:95) of 96.1% during cross validation, making it as the most effective model for generalizing to unseen data.
Cloud environments built on microservices offer great scalability and flexibility, but they are increasingly vulnerable to advanced Denial-of-Service (DoS) attacks. In emerging 6G networks, where services are distributed and resources are dynamically allocated, these threats become even more difficult to manage. The combination of multiple entry points, tight inter-service links, and features like cognitive radio and AI-driven spectrum sensing further amplifies the risks. This paper proposes a graph-aware anomaly detection framework that models the cloud environment as a directed endpoint graph, aggregates flow-level features into windowed vectors, and characterizes legitimate and attack traffic using non-homogeneous Poisson processes and Hurst exponent analysis. An exponentially weighted moving average (EWMA) deviation score with graph-based contextual boosting is combined with logistic regression to enhance detection accuracy, supporting optimal resource management and interference mitigation in 6G multi-connectivity architectures. The framework is evaluated on the recent BCCC-cPacket-Cloud-DDoS-2024 dataset, achieving consistently lower false positive rates and higher detection rates than Gaussian Naive Bayes, with a peak F1-score of 0.93.
A dual-pattern reconfigurable antenna comprising a torus-shaped metasurface controlled by flip-chip Aluminum Gallium Arsenide (AlGaAs) PIN diodes and a bicone antenna feed is investigated. A broadband bicone antenna operating between 14 and 34 GHz is initially optimized to enhance its realized gain and achieve 4.1 dBi at 29 GHz and 4.4 dBi at 30 GHz. The designed bicone is subsequently used as a feed for the torus-shaped metasurface. A torus-shaped metasurface capable of switching between reflection and transmission is investigated in two stages. Its geometry is optimized as a metallic reflector to reduce computational time. The dimensions of the reflector are optimized to improve the gains and lower the side lobe levels (SLL). The simulation of the reflector confirms that the torus reflector with the bicone feed achieves a realized gain of 16.3 dBi at 29 GHz and 17 dBi at 30 GHz. The proposed metasurface achieves transmission and reflection switching between 27.5 and 30.5 GHz by controlling the PIN diode state. Based on the results, the proposed system can operate in two distinct modes: omnidirectional and directional. The expected omnidirectional performance of the final system corresponds to that of the standalone bicone antenna, reduced by the metasurface insertion loss. In the directional mode, the realized gain and radiation pattern are determined by the combined response of the bicone antenna and torus-shaped metasurface reflector.
This paper explores the application of Game Theory techniques to Downlink-Uplink Decoupling (DUDe) in 5G networks, with the goal of optimizing resource allocation. DUDe technology enables the independent management of uplink and downlink connections, introducing new challenges in the distribution of network resources such as bandwidth and energy. To address these challenges, the study investigates two specific game-theoretic algorithms: the Gale-Shapley algorithm, which models a matching game between users and base stations, and the Nash Bargaining algorithm, which dynamically adjusts bandwidth allocation based on individual user demands. By applying these algorithms, the paper demonstrates how Game Theory can offer effective and adaptive solutions to complex resource management problems in next-generation networks. The results highlight improvements in network performance, energy efficiency, and user experience. This work underscores the relevance of Game Theory in the design of intelligent 5G infrastructures and provides a foundation for future research in autonomous resource allocation in distributed network environments.
This paper proposes a data rate enhancement scheme for frequency-modulated continuous wave (FMCW) radar systems combining ranging and communication. The conventional FMCW radars use continuous and linearly sweeping chirp signals. In the proposed method, data are modulated onto the sweep function of the FMCW radar. The chirp signal for each symbol is divided into time-slots, and multiple sweep functions are generated by applying frequency hopping to both up-chirp and down-chirp waveforms. By assigning different sweep functions, the system can transmit more data per symbol. At the receiver, the beat frequency is calculated from the reflected received signal and a reference signal, enabling distance estimation as in conventional beat frequency ranging methods. The performance of the proposed scheme is evaluated through simulations in terms of root mean square error (RMSE) and bit error rate (BER), and is compared with that of conventional FMCW radar.
