Related papers: Time-drift Aware RF Optimization with Machine Lear…
The LIGHT (Linac for Image-Guided Hadron Therapy) project was initiated to develop a modular proton accelerator delivering beam with energies up to 230 MeV for cancer therapy. The machine consists of three different kinds of accelerating…
Modern machine learning is trained by stochastic gradient descent (SGD), whose performance critically depends on how the learning rate (LR) is adjusted and decreased over time. Yet existing LR regimes may be intricate, or need to tune one…
A light source based on an Energy Recovered Linac (ERL) [1] consist of a superconducting linac and a transfer line that includes wigglers and undulators to produce the synchrotron light. The transfer line brings the electrons bunches back…
Intelligent reflecting surface (IRS) has been recently employed to reshape the wireless channels by controlling individual scattering elements' phase shifts, namely, passive beamforming. Due to the large size of scattering elements, the…
Unforeseen particle accelerator interruptions, also known as interlocks, lead to abrupt operational changes despite being necessary safety measures. These may result in substantial loss of beam time and perhaps even equipment damage. We…
Several research works have applied Reinforcement Learning (RL) algorithms to solve the Rate Adaptation (RA) problem in Wi-Fi networks. The dynamic nature of the radio link requires the algorithms to be responsive to changes in link…
The development of Ultra-High Frequency (UHF) linear accelerators via Metal Additive Manufacturing (MAM) is a strategic research focus of the RACERS team at GSI. The 704.4 MHz Crossbar H-mode (CH) cavity, proposed in 2021 to facilitate…
This paper proposes Load-aware Tram-FL, an extension of Tram-FL that introduces a training scheduling mechanism to minimize total training time in decentralized federated learning by accounting for both computational and communication…
Machine learning (ML) is a subfield of artificial intelligence. The term applies broadly to a collection of computational algorithms and techniques that train systems from raw data rather than a priori models. ML techniques are now…
This paper addresses the challenges of wideband signal beamforming in radar systems and proposes a new calibration method. Due to operating conditions, the frequency dependent characteristics of the system can be changed, and amplitude,…
The increasing deployment of advanced digital technologies such as Internet of Things (IoT) devices and Cyber-Physical Systems (CPS) in industrial environments is enabling the productive use of machine learning (ML) algorithms in the…
Beam selection for millimeter-wave links in a vehicular scenario is a challenging problem, as an exhaustive search among all candidate beam pairs cannot be assuredly completed within short contact times. We solve this problem via a novel…
With the new GSI High Current Injector, the beam pulse intensity will be increased by more than two orders of magnitude. The high beam power and the short stopping range at particle energies below 12 MeV/u can destroy accelerator components…
Integrated Sensing and Communication (ISAC) is a key enabler in 6G networks, where sensing and communication capabilities are designed to complement and enhance each other. One of the main challenges in ISAC lies in resource allocation,…
The phase averaging reference line system provides the RF phase reference, LO and clock signals to the LLRF and other accelerator sub-systems. The PIP-II linac has RF systems at three frequencies - 162.5 MHz, 325 MHz and 650 MHz. A…
Diffusion models have recently achieved remarkable success in generative modeling, yet their training dynamics across different noise levels remain highly imbalanced, which can lead to inefficient optimization and unstable learning…
The Fermilab High Intensity Neutrino Source program has built and commissioned a pulsed 325 MHz RFQ. The RFQ has successfully accelerated a proton beam at the RFQ design RF power. Experiences encountered during RFQ conditioning, including…
It is important to have experimental methods to estimate the maximum beam intensity for the Fermilab Booster as objective input into long term program commitments. An important existing limit is set by the available rf power. This limit is…
Recent breakthroughs in Deep Learning (DL) applications have made DL models a key component in almost every modern computing system. The increased popularity of DL applications deployed on a wide-spectrum of platforms have resulted in a…
Beam breakup instability is a potential issue for all particle accelerators and is often the limiting factor for the maximum beam current that can be achieved. This is particularly relevant for Energy Recovery Linacs with multiple passes…