arXiv subjects

Holger Schlarb

Publications and source records attributed to Holger Schlarb.

15 recordsLinked to original sources

arXiv2026

Ageing Monitoring for Commercial Microcontrollers Based on Timing Windows

Microcontrollers are increasingly present in embedded deployments and dependable systems, for which malfunctions due to hardware ageing can have severe impact. The lack of deployable techniques for ageing monitoring on these devices has spread the application of guard bands to prevent timing errors due to degradation. Applying this static technique can limit performance and lead to sudden failures as devices age. In this paper, we follow a software-based self-testing approach to design monitoring of hardware degradation for microcontrollers. Deployable in the field, our technique leverages timing windows of variable lengths to determine the maximum operational frequency of the devices. We empirically validate the method on real hardware and find that it consistently detects temperature-induced degradations in maximum operating frequency of up to 13.79 % across devices for 60 °C temperature increase.

cs.AR↗

arXiv2025

Distributed Acoustic Fiber Sensing for Research Campuses and Large Scientific Infrastructures -- The Hamburg WAVE proto-network

Here, we demonstrate and investigate how Distributed Acoustic Sensing (DAS) can be utilized on research campuses and in large scientific infrastructures to study environmental vibrations and reduce their impact on high-precision experiments. We first discuss the potential of DAS in the context of particle accelerators, gravitational wave detection experiments and research campuses. Next, we present the results of our seismic measurement campaign conducted with our proto-network, which involved the probing of over 12 km of fiber, in May 2021. This campaign was conducted by the Hamburg WAVE initiative in Science City Hamburg Bahrenfeld and included DESY, the European XFEL, PETRA III and the University of Hamburg. Our proto-network confirms the ability to observe natural, anthropogenic, and infrastructural vibrations and how and where these couple into different parts of the heterogeneously set up fiber network. We also present results on a study of noise and motion coupling aspects of DAS probing double-redundant fiber loops in a unique environment, the European XFEL. Our results show that DAS greatly benefits research campuses and large scientific infrastructures and they highlight the opportunities and challenges of implementing and operating such seismic networks.

physics.ins-det↗

arXiv2025

Homogenized harmonic balance finite element method for nonlinear eddy current simulations of fast corrector magnets

This paper develops a homogenized harmonic balance finite element method (HomHBFEM) to predict the dynamic behavior of magnets with fast excitation cycles, including eddy current and skin effects. A homogenization technique for laminated yokes avoids resolving the individual laminates and the skin depth in the finite element (FE) mesh. Instead, the yoke is represented by a bulk surrogate material with frequency-dependent parameters. The ferromagnetic saturation of the yoke at higher excitation currents is tackled by a harmonic balance method, which accounts for a coupled set of frequency components. Thereby, a computationally expensive time-stepping of the eddy-current field problem and a convolution of the homogenized yoke model are avoided. The HomHBFEM enables, for the first time, to conduct nonlinear simulations of fast corrector magnets, which are embedded in a fast orbit feedback system to counteract orbit disturbances over a broad frequency spectrum, and thus guarantee a stable light-source operation. The results show the impact of the nonlinearity on the phase lag and the field attenuation as well as the eddy current losses at frequencies up to 65 kHz. The numerical validation for a C-dipole magnet example shows that the HomHBFEM achieves a sufficient accuracy at an affordable computational effort, with simulation times of a few hours. In comparison, standard 3D transient FE simulations need to resolve the lamination thickness and the skin depth in space and the largest relevant frequency in time, which leads to a two to three orders of magnitude larger mesh and prohibitive computational effort, with simulation times of a few weeks on a contemporary computer server.

physics.acc-ph↗

arXiv2025

Impact of Seismic Activities on the Optical Synchronization System of the European X-ray Free-Electron Laser

This study investigates the influence of seismic activities on the optical synchronization system of the European X-ray Free-Electron Laser. We analyze the controller I/O data of phase-locked-loops in length-stabilized links, focusing on the response to earthquakes, ocean-generated microseism and civilization noise. By comparing the controller data with external data, we were able to identify disturbances and their effects on the control signals. Our results show that seismic events influence the stability of the phase-locked loops. Even earthquakes that are approximately \qty{5000}{\km} away cause remarkable fluctuations in the in-loop control signals. Ocean-generated microseism in particular has an enormous influence on the in-loop control signals due to its constant presence. The optical synchronization system is so highly sensitive that it can even identify vibrations caused by civilization, such as road traffic or major events like concerts or sport events. The phase-locked loops manages to eliminate more than 99% of the existing interference.

hep-ex↗

arXiv2024

Switching Frequency as FPGA Monitor: Studying Degradation and Ageing Prognosis at Large Scale

The growing deployment of unhardened embedded devices in critical systems demands the monitoring of hardware ageing as part of predictive maintenance. In this paper, we study degradation on a large deployment of 298 naturally aged FPGAs operating in the European XFEL particle accelerator. We base our statistical analyses on 280 days of in-field measurements and find a generalized and continuous degradation of the switching frequency across all devices with a median value of 0.064%. The large scale of this study allows us to localize areas of the deployed FPGAs that are highly impacted by degradation. Moreover, by training machine learning models on the collected data, we are able to forecast future trends of frequency degradation with horizons of 60 days and relative errors as little as 0.002% over an evaluation period of 100 days.

cs.AR↗

arXiv2024

A Review of Techniques for Ageing Detection and Monitoring on Embedded Systems

Embedded digital devices are progressively deployed in dependable or safety-critical systems. These devices undergo significant hardware ageing, particularly in harsh environments. This increases their likelihood of failure. It is crucial to understand ageing processes and to detect hardware degradation early for guaranteeing system dependability. In this survey, we review the core ageing mechanisms, identify and categorize general working principles of ageing detection and monitoring techniques for Commercial-Off-The-Shelf (COTS) components that are prevalent in embedded systems: Field Programmable Gate Arrays (FPGAs), microcontrollers, System-on-Chips (SoCs), and their power supplies. From our review, we find that online techniques are more widely applied on FPGAs than on other components, and see a rising trend towards machine learning application for analysing hardware ageing. Based on the reviewed literature, we identify research opportunities and potential directions of interest in the field. With this work, we intend to facilitate future research by systematically presenting all main approaches in a concise way.

cs.AR↗

arXiv2024

Studying the Degradation of Propagation Delay on FPGAs at the European XFEL

An increasing number of unhardened commercial-off-the-shelf embedded devices are deployed under harsh operating conditions and in highly-dependable systems. Due to the mechanisms of hardware degradation that affect these devices, ageing detection and monitoring are crucial to prevent critical failures. In this paper, we empirically study the propagation delay of 298 naturally-aged FPGA devices that are deployed in the European XFEL particle accelerator. Based on in-field measurements, we find that operational devices show significantly slower switching frequencies than unused chips, and that increased gamma and neutron radiation doses correlate with increased hardware degradation. Furthermore, we demonstrate the feasibility of developing machine learning models that estimate the switching frequencies of the devices based on historical and environmental data.

cs.AR↗

arXiv2023

Ageing Analysis of Embedded SRAM on a Large-Scale Testbed Using Machine Learning

Ageing detection and failure prediction are essential in many Internet of Things (IoT) deployments, which operate huge quantities of embedded devices unattended in the field for years. In this paper, we present a large-scale empirical analysis of natural SRAM wear-out using 154 boards from a general-purpose testbed. Starting from SRAM initialization bias, which each node can easily collect at startup, we apply various metrics for feature extraction and experiment with common machine learning methods to predict the age of operation for this node. Our findings indicate that even though ageing impacts are subtle, our indicators can well estimate usage times with an $R^2$ score of 0.77 and a mean error of 24% using regressors, and with an F1 score above 0.6 for classifiers applying a six-months resolution.

cs.AR↗

arXiv2023

Learning to Do or Learning While Doing: Reinforcement Learning and Bayesian Optimisation for Online Continuous Tuning

Online tuning of real-world plants is a complex optimisation problem that continues to require manual intervention by experienced human operators. Autonomous tuning is a rapidly expanding field of research, where learning-based methods, such as Reinforcement Learning-trained Optimisation (RLO) and Bayesian optimisation (BO), hold great promise for achieving outstanding plant performance and reducing tuning times. Which algorithm to choose in different scenarios, however, remains an open question. Here we present a comparative study using a routine task in a real particle accelerator as an example, showing that RLO generally outperforms BO, but is not always the best choice. Based on the study's results, we provide a clear set of criteria to guide the choice of algorithm for a given tuning task. These can ease the adoption of learning-based autonomous tuning solutions to the operation of complex real-world plants, ultimately improving the availability and pushing the limits of operability of these facilities, thereby enabling scientific and engineering advancements.

cs.LG↗

arXiv2022

RF regulation with superconducting cavities and beam operation using a frequency shifted cavity

The free-electron laser FLASH at DESY and the European XFEL are operated with superconducting radio frequency cavities and supply beam to several user experiments. The switching time between experiments is limited to dozens of microseconds. This contribution will show a regulation with a frequency shifted superconducting cavity to manipulate and change the accelerating properties of electron bunches with 250 kHz.The main challenge of the concept presented in this contribution can be summarized in this statement: finding a way to modulate the energy of individual bunches in a single-source multiple-cavity scheme for a potential CW upgrade of the EuXFEL.

physics.acc-ph↗

arXiv2021

High-fidelity Prediction of Megapixel Longitudinal Phase-space Images of Electron Beams using Encoder-Decoder Neural Networks

Modeling of large-scale research facilities is extremely challenging due to complex physical processes and engineering problems. Here, we adopt a data-driven approach to model the longitudinal phase-space diagnostic beamline at the photoinector of the European XFEL with an encoder-decoder neural network model. A deep convolutional neural network (decoder) is used to build images measured on the screen from a small feature map generated by another neural network (encoder). We demonstrate that the model trained only with experimental data can make high-fidelity predictions of megapixel images for the longitudinal phase-space measurement without any prior knowledge of photoinjectors and electron beams. The prediction significantly outperforms existing methods. We also show the scalability and interpretability of the model by sharing the same decoder with more than one encoder used for different setups of the photoinjector, and propose a pragmatic way to model a facility with various diagnostics and working points. This opens the door to a new way of accurately modeling a photoinjector using neural networks and experimental data. The approach can possibly be extended to the whole accelerator and even other types of scientific facilities.

cs.LG↗

arXiv2021

Pickup concepts for ultra-low charged short bunches in X-Ray Free-Electron Lasers

The all-optical synchronization systems used in various X-ray free-electron lasers (XFEL) such as the European XFEL observe the transient fields of passing electron bunches coupled into one or more pickups in the Bunch Arrival Time Monitors (BAM). The extracted signal is then amplitude modulated on reference laser pulses in a Mach-Zehnder type electro-optical modulator. With the emerging demand for future experiments with ultra-short FEL shots, fs precision is required for the synchronization systems even with 1 pC bunches. Since the sensitivity of the BAM depends in particular on the slope of the bipolar signal at the zero-crossing and thus, also on the bunch charge, a redesign with the aim of a significant increase by optimized geometry and bandwidth is inevitable. In this contribution the theoretical foundations of the pickup signal are aggregated and treated with a focus on ultra-short bunches as well as a general formulation. A possible new pickup concept is simulated and its performance is compared to the previous concept. A significant improvement of slope and voltage is found. The improvement is mainly achieved by the reduced distance to the beam and a higher bandwidth.

physics.ins-det↗

arXiv2018

Real-Time Redundancy for the 1.3 GHz Master Oscillator of the European-XFEL

Many modern large-scale facilities, like European X-ray Free Electron Laser (E-XFEL), require precise synchronisation, often down to femtosecond level. Even a very short interruption or an excessive glitch in the reference signal might break the precise time relations between subsystems. In such event, a time-consuming resynchronization process is required that renders the facility not available for the users until it is completed. Therefore, such events are highly undesirable. In this paper, we present an autonomous redundancy solution for the European-XFEL's master oscillator that will guarantee a continuous delivery of the high-quality reference signal even in case of most of the potential failures. The concept and implementation are presented, as well as results from testing in the laboratory environment.

physics.ins-det↗

arXiv2018

Phase Drift Compensating RF Link for Femtosecond Synchronization of E-XFEL

Modern high-energy particle accelerators and Free-Electron Lasers incorporate large quantities of sensitive RF and microwave frequency devices distributed over kilometer distances. Such devices require extreme stable phase and time synchronization by means of high frequency signal distributed along the accelerator facility. Coaxial cables are commonly used to distribute the reference signal over the large machine to synchronize electronic systems and they are the main source of undesirable phase drifts in the synchronization system. Signal phase drifts in cables are mainly caused by temperature and humidity variations and their values usually exceed required phase synchronization accuracy by more than order of magnitude. There are several approaches to reduce signal phase drifts in coaxial cables. This paper describes the realization of active phase stabilization system based on interference phenomenon. A phase-locked signal from the transmitter is reflected at the end of a coaxial cable link. Directional couplers placed along the cable pick up the forward and reflected signals and interfere them to cancel out the cable phase drifts. Distributed hardware including interferometer controller/transmitter and receiver modules were built to demonstrate system concept and performance. Link input and output devices used FPGA I/O boards with Ethernet interface to control system operation. Specialized firmware and software was developed to calibrate and control the system. This paper describes the concept of interferometer link, designed hardware, basic control algorithms and performance evaluation results. The link prototype was built to distribute 1.3 GHz signal through a coaxial cable. Measured phase drift suppression factor value exceeded level of 100.

physics.ins-det↗

arXiv2012

High Bandwidth Pickup Design for Bunch Arrival-time Monitors for Free-Electron Laser

In this paper, we present the design and realization of high bandwidth pickup electrodes with a cutoff frequency above 40 GHz. The proposed cone-shaped pickups are part of a bunch arrival-time monitor (BAM) designed for high (> 500 pC) and low (20 pC) bunch charge operation mode providing for a time resolution of less than 10 fs for both operation modes. The proposed design has a fast voltage response, low ringing, and a resonance-free spectrum. For assessing the influence of manufacturing tolerances on the performance of the pickups, an extensive tolerance study has been performed via numerical simulations. A non-hermetic model of the pickups was built for measurement and validation purposes. The measurement and simulation results are in good agreement and demonstrate the capability of the proposed pickup system to meet the given specifications.

physics.acc-ph↗