Machine learning algorithms for triangulation-based optical beam tracking in FSO systems

With the rapid deployment of high-speed maglev transportation systems worldwide, the operational velocity, electromagnetic complexity, and channel dynamics have far exceeded those of conventional rail systems, imposing more stringent requirements on real-time capability, reliability, and interference robustness in wireless communication. In maglev environments exceeding 600 km/h, the channel becomes predominantly line-of-sight with sparse scatterers, exhibiting strong Doppler shifts, rapidly varying spatial characteristics, and severe interference, all of which significantly degrade the stability and convergence performance of traditional beamforming algorithms. Adaptive smart antenna technology has therefore become essential in high-mobility communication and sensing systems, as it enables real-time spatial filtering, interference suppression, and beam tracking through continuous weight updates. To address the challenges of slow convergence and high steady-state error in rapidly varying maglev channels, this work proposes a new Fractional Proportionate Normalized Least Mean Square (FPNLMS) adaptive beamforming algorithm. The contributions of this study are twofold. (1) A novel FPNLMS algorithm is developed by embedding a fractional-order gradient correction into the power-normalized and proportionate gain framework of PNLMS, forming a unified LMS-type update mechanism that enhances error tracking flexibility while maintaining O(L) computational complexity. This integrated design enables the proposed method to achieve faster convergence, improved robustness, and reduced steady-state error in highly dynamic channel conditions. (2) A unified convergence analysis framework is established for the proposed algorithm. Mean convergence conditions and practical step-size bounds are derived, explicitly incorporating the fractional-order term and generalizing classical LMS/PNLMS convergence theory, thereby providing theoretical guarantees for stable deployment in high-speed maglev beamforming. Simulation results verify that the proposed FPNLMS algorithm achieves significantly faster convergence, lower mean square error, and superior interference suppression compared with LMS, NLMS, FLMS, and PNLMS, demonstrating its strong applicability to beamforming in highly dynamic next-generation maglev communication systems.

Abstract Reconfigurable Intelligent Surfaces (RISs) are a key enabler for 6G networks, enhancing spectral efficiency and reliability in non‐line‐of‐sight channels. However, user mobility introduces beam misalignment and phase tracking challenges. This paper proposes the group reflection–based RIS with beam tracking (GR‐RIS‐BT) framework, which integrates group reflection modulation (GRM) with unscented Kalman filter (UKF)–based tracking. The transmitted data are divided into two streams: one configures RIS element groups, while the other conveys modulated symbols. This bit‐splitting mechanism doubles spectral efficiency without significantly increasing complexity. Simulation results show that GR‐RIS‐BT achieves a bit error rate (BER) of 10 −4 at 15 dB SNR using a four‐group RIS, outperforming conventional RIS schemes without beam tracking. Moreover, UKF‐based tracking ensures robust phase alignment under mobility, sustaining low error rates and reliable connectivity. The proposed GR‐RIS‐BT framework offers a scalable and efficient solution for high‐mobility 6G systems by jointly optimizing spectral efficiency, BER, and computational complexity.

In satellite laser communications, maintaining stable beam tracking is a prerequisite for establishing reliable optical links. To overcome relative motion between terminals, platform vibration, and other external disturbances, the pointing, acquisition, and tracking (PAT) system typically employs a composite axis control structure, which utilizes sensors to detect deviations between incident light and the optical axis, then corrects the pointing of the optical antenna through the drive unit. To address issues such as the low dynamic response characteristics of coarse tracking systems, receiving optical signal detection delays, closed-loop control lag, and coarse-fine tracking cascade delays, this paper proposes a hybrid tracking method based on long short-term memory (LSTM) neural networks for terminal pointing prediction. During the tracking phase, LSTM uses historical pointing trajectories to accomplish short-term pointing predictions for the terminal. Compared to using ephemeris forecasts, this method is not affected by factors such as orbital prediction errors and satellite attitude control errors, and does not introduce additional systematic errors. The integration of the pointing prediction mechanism and single-detector closed-loop tracking improves the dynamic tracking capability and robust stability of the PAT system. Simulation and experiments verified the effectiveness of the proposed method.

This study investigates the communication network (MUAVN) of intelligent reflecting surface (IRS)-assisted high-speed multiple unmanned aerial vehicles, considering that highly dynamic UAVs may incur poor performance due to severe channel fading and rapid channel changes. Our objective is to design an adaptive dual-beam tracking scheme that mitigates beam misalignment, enhances the performance of the worst-case UAV, and sustains reliable communication links in the high-speed MUAVNs (HSMUAVNs). We first exploit an attention-based double-layer long short-term memory network to predict the spatial angle information of each UAV, which yields optimal beam coverage that matches to the UAV’s actual flight trajectory. Then, a worst-case UAV’s received beam components signal-to-interference plus noise ratio (SINR) maximization problem is formulated by jointly optimizing ground base station’s beam components and IRS’s phase shift matrix. To address this challenging problem, we decouple the optimization problem into two subproblems, which are then solved by leveraging semi-definite relaxation, the bisection method, and eigenvalue decomposition techniques. Finally, the adaptive dual beams are generated by linearly weighting the obtained beam components, each of which is well-matched to the corresponding moving UAV. Numerical results reveal that the proposed beam tracking scheme not only enhances the worst-case UAV’s performance but also guarantees a sufficient SINR demanded across the entire HSMUAVN.

Artificial Intelligence-Aided Beam Tracking in Autonomous Vehicles: State of the Art and Future Directions

Adaptive optical beam tracking and alignment system with wide field-of-view (FOV) for optical wireless communication (OWC)

A high-speed receiver module for free-space optical communications with robust alignment, incorporating a previously developed photodetector array (PDA), is presented. The packaged receiver was characterized for the frequency response and transmission performance. By optimizing the PCB on which the array is mounted, a flatter modulation response and higher element bandwidth were achieved. The receiver module exhibited an average 3 dB bandwidth of 2.7 GHz, with an alignment tolerance of 37 µm without mechanical beam tracking. The 3 dB, 6 dB, and 10 dB bandwidth have all significantly improved compared with prior work. In a 25 km SMF transmission demonstration, 10 selected elements operated reliably at 10 Gbps (NRZ). With additional transmitter side equalization, NRZ bit error rates (BERs) below 10-12 were achieved at up to 10 Gbps.

It is of paramount importance to accurately estimate the angular direction of the optical beam incident at the receiver for fast and precise beam alignment between the transmitter and receiver in free-space optical communications. In this work, we propose a beam position estimation method for a quadrant detector and experimentally demonstrate fast beam tracking using the proposed method. The experimental results show that the proposed method tracks the beam 21% and 36% faster than the conventional difference-over-sum and log-ratio methods, respectively.

Free-space optical (FSO) communication is a promising technology for high-speed data transmission, but its effectiveness is highly dependent on precise beam alignment. In this work, we present a computer vision-assisted tracking system designed to maintain robust optical alignment in real time. By combining a lightweight convolutional neural network (CNN) with a Kalman filter, the system can detect the laser spot accurately and adjust the beam direction through a closed-loop feedback mechanism. Our experimental results show 98.5% tracking accuracy and reliable data transmission at 1 Gbps over distances up to 2 km. The system performs consistently in a variety of conditions, including fog, wind, motion blur, and glare. It significantly reduces bit error rates and improves signal stability compared to conventional tracking approaches. Running on an embedded Jetson Xavier NX platform, the system achieves low-latency operation and efficient power consumption, making it suitable for UAV and satellite applications. These results demonstrate the practical advantages of integrating computer vision into optical communication systems, especially where fast, accurate, and adaptive beam alignment is required. Future work will explore predictive tracking, multi-sensor fusion, and adaptive modulation to further improve performance in extreme conditions.

Underwater wireless laser communication (UWLC), as a key technology in deep-sea exploration and underwater operations, has attracted the attention of many researchers. Maintaining stable and high-data-rate UWLC has been eagerly pursued. However, in the practical implementation of UWLC, underwater disturbances, flow and waves may inevitably disable the operation of the underwater communication link. Normally, the underwater disturbances can take place as fast as the millisecond time scale. Therefore, fast closed loop tracking technologies for stable UWLC links are necessary to mitigate the underwater interference. In this paper, we propose a fast beam tracking scheme based on a dynamic vision sensor (DVS), for real-time fast tracking for UWLC links when encountering underwater disturbances. By combining a high-time-resolution DVS with a proportional-integral-derivative (PID) control algorithm, the beam tracking system can accurately detect the centroid offset of light spots with a time resolution of <1 ms, which is the best temporal resolution for underwater beam acquisition to the best of our knowledge. Real-time beam alignment through a 11-ms closed loop control is achieved for UWLC links. This experimentally validates that the combination of the high temporal resolution of DVS and PID algorithms is able to deliver a simple and fast tracking method, which may be critical for future breakthroughs in tracking technologies for UWLC links.

To address the issue of insufficient contrast in conventional X-ray absorption imaging for biological soft tissues and weakly absorbing materials, this paper proposes a beam tracking X-ray phase-contrast imaging system using a conventional X-ray source. A periodic pinhole array mask is placed between the X-ray source and the sample to spatially modulate the X-ray beam, dividing it into multiple independent sub-beams. Each sub-beam is deflected due to the modulation effect of the sample, resulting in slight positional shifts in the intensity patterns formed on the detector. The experiments employ an X-ray source with a 400 μm focal spot and use a two-dimensional step-scanning approach to acquire image sequences of various samples. The experimental results show that this method can extract the edge profile and structural changes in the samples to some extent, and it demonstrates good contrast and detail recovery under weak absorption conditions. These results suggest that this method has certain application potential in material inspection, non-destructive testing, and related fields.

Analyzing Cardiorespiratory Motion and Its Dosimetric Effect on Stereotactic Arrhythmia Radio-Ablation: A STOPSTORM.eu Consortium Study.

Demonstration of Optoelectronic THz Beam Tracking Using Real-Time Phase Optimization

SACLA is the world’s most compact hard x-ray free-electron laser and has been successfully and reliably operated for over a decade. To meet increasing user demands, an upgrade of the SACLA linear accelerator is planned in order to achieve a kHz-level repetition rate and enhanced beam brightness. A key component of this upgrade is a new injector, based on the existing pulsed dc gun with a thermionic cathode, which has demonstrated high beam quality, excellent stability, and simple maintenance. In this work, we present an in-depth beam dynamics study of a highly promising new SACLA injector architecture. Multistage velocity bunching, combined with one stage of magnetic compression and careful control of emittance growth, allows for up to 3 orders of magnitude of bunch compression while keeping the emittance growth at the submicron level. Our optimization approach utilizes a genetic algorithm coupled to 3D space-charge-dominated beam tracking to refine beam performance and identify optimal injector parameters. Thus, we demonstrate that the proposed injector can deliver electron beams with bunch durations and emittances comparable to those produced by the state-of-the-art C-band injector based on an rf photocathode gun [A. Giribono , ]. Additionally, our extensive beam dynamics analysis of the dc electron gun (see part 2) provides valuable insights applicable to low-voltage continuous-wave very high frequency guns and superconducting radio-frequency guns.

Beam Tracking for High-Speed UAV via Generative Diffusion Model-Enabled Joint Optimization Approach

Programmable metasurfaces have emerged as one of the most promising technologies for next‐generation wireless communication due to their powerful capabilities of dynamically controlling electromagnetic waves. Here, an intelligent wireless communication system is proposed, assisted by programmable metasurfaces and computer‐vision to achieve self‐adaptive signal enhancement of the moving users. In this system, high‐speed adaptive beam tracking for moving users is achieved, which recognizes users’ posture and detects the position of their hands by a monocular camera through a computer‐vision algorithm based on a convolutional neural network. A prototype is fabricated and measured indoors and outdoors, which demonstrates that, without human intervention, the received power and signal‐to‐interference‐plus‐noise ratio of the moving user with different moving speeds can be significantly enhanced. The proposed system is expected to promote the development and application of programmable metasurfaces in future intelligent wireless communication.

This study investigates a large-scale dynamic Vehicle-to-Everything (V2X) communication network, in which multiple Roadside Units (RSUs) are deployed along highways to enable high-speed vehicular links. To ensure robust and adaptive performance under fast-varying conditions, we propose an integrated framework that combines resource block-based MC-CDMA modulation with dynamic beamforming optimized for complex propagation environments. A custom code mapper and resource element (RE) allocator are introduced to support interference-aware transmission and enhance signal robustness in dense deployment scenarios. The MC-CDMA scheme enables extended-range coverage per RSU, outperforming traditional OFDM-based transmission in terms of reliability and scalability. To further optimize performance, a Deep Reinforcement Learning (DRL) model is employed to jointly handle beam tracking and time-varying channel conditions. Specifically, a physics-inspired Deep Q-Learning (DQL) strategy is proposed, using a force-arm-based mechanism to adaptively correct beam misalignment caused by mobility and Doppler effects. Simulation results demonstrate that the proposed system achieves significant improvements in bit error rate (BER), bitrate stability, handover smoothness, and spectral efficiency. When equipped with a large-scale antenna array, the system ensures continuous beam tracking and substantially outperforms conventional RL-based techniques. These results highlight its potential for future 6G-enabled V2X deployments, where scalability, adaptability, and robust link quality are essential.

The rapid development of metasurfaces offers new possibilities to establish novel wireless communication systems with simplified architectures. However, the current demonstration systems are based on the reflection-type metasurfaces, which suffer from high profiles and integration challenges in practice. Such configurations are also inefficient for handling multiple subcarriers during beam scanning and beam tracking. To address these limitations, a radiation-type metasurface fed by a microstrip array antenna is proposed in this paper, which is used to construct a new-architecture wireless communication system. Compared to the reported metasurface-based communication systems, the proposed design is more flexible for information modulation and transmission, with the system profile significantly reduced. The phase modulation is implemented by changing the transmission phase of metasurface, allowing for baseband signals to be directly imparted to the carrier wave from the feeding source. A real-time signal transmission experiment validates the performance of the proposed metasurface-based communication system.

Two complementary processes of radio antennamain narrow beam pointing to the selected communications satellite are considered: the initial angular beam setting in the direction of the satellite and its beam tracking in order to improve the conditions of radio communication. A rotary support device with control by two angular coordinates: azimuth angle and altitude angle,was chosen as a platform to mount the antenna. At the same time, the rotary support mounting can be carried out both on a resting and on a moving ground platform. The geostationary and low-orbit satellite options were chosen as the communications satellite. As a result, calculation formulae have been obtained and discussed to ensure the following important and frequently encountered in practice modes of ground antennapointing at communications satellites with a circular orbit: setting the angular direction of the ground antenna main beam to a geostationary communications satellite; pointing the transported antenna main beam to the geostationary communication satellite; initial exposure of the angular direction of a stationary ground antenna beam for the acquisition of a low-orbit communications satellite; pointing the main beam of a stationary ground antenna at a moving low-orbit communications satellite; pointing the transported antenna main beam to a low-orbit communications satellite. At the same time, the known dependencies corresponding to the mode of non-guidance of the antenna support and rotary device on the geostationary communications satellite were chosen as the basic mathematical model. The introduction of appropriate substitutions made it possible to be generalized to all other listed pointing modes. The article provides some illustrative calculationsto get an initial idea of the requirements for the pointing system. In general, the disclosure of the articlematerial is carried out from the point of view of a rotary support device developer intereststo implement effective antenna pointing at a communications satellite at the angles of azimuth and altitude.