| 1. |
- Iqbal, Muhammad, et al.
(författare)
-
Symbol error rate minimization using deep learning approaches for short-reach optical communication networks
- 2024
-
Ingår i: Frontiers in Physics. - : Frontiers Media S.A.. - 2296-424X. ; 12
-
Tidskriftsartikel (refereegranskat)abstract
- Short-reach optical communication networks have various applications in areas where high-speed connectivity is needed, for example, inter- and intra-data center links, optical access networks, and indoor and in-building communication systems. Machine learning (ML) approaches provide a key solution for numerous challenging situations due to their robust decision-making, problem-solving, and pattern-recognition abilities. In this work, our focus is on utilizing deep learning models to minimize symbol error rates in short-reach optical communication setups. Various channel impairments, such as nonlinearity, chromatic dispersion (CD), and attenuation, are accurately modeled. Initially, we address the challenge of modeling a nonlinear channel. Consequently, we harness a deep learning model called autoencoders (AEs) to facilitate communication over nonlinear channels. Furthermore, we investigate how the inclusion of a nonlinear channel within an autoencoder influences the received constellation as the optical fiber length increases. Another facet of our work involves the deployment of a deep neural network-based receiver utilizing a channel influenced by chromatic dispersion. By gradually extending the optical length, we explore its impact on the received constellation and, consequently, the symbol error rate. Finally, we incorporate the split-step Fourier method (SSFM) to emulate the combined effects of nonlinearities, chromatic dispersion, and attenuation in the optical channel. This is accomplished through a neural network-based receiver. The outcome of this work is an evaluation and reduction of the symbol error rate as the length of the optical fiber is varied.
|
|
| 2. |
- Pennicard, David, et al.
(författare)
-
Data reduction and processing for photon science detectors
- 2024
-
Ingår i: Frontiers in Physics. - : Frontiers Media SA. - 2296-424X. ; 12
-
Forskningsöversikt (refereegranskat)abstract
- New detectors in photon science experiments produce rapidly-growing volumes of data. For detector developers, this poses two challenges; firstly, raw data streams from detectors must be converted to meaningful images at ever-higher rates, and secondly, there is an increasing need for data reduction relatively early in the data processing chain. An overview of data correction and reduction is presented, with an emphasis on how different data reduction methods apply to different experiments in photon science. These methods can be implemented in different hardware (e.g., CPU, GPU or FPGA) and in different stages of a detector's data acquisition chain; the strengths and weaknesses of these different approaches are discussed.
|
|
| 3. |
- Suresh, Rahul, et al.
(författare)
-
Revolutionizing physics : a comprehensive survey of machine learning applications
- 2024
-
Ingår i: Frontiers in Physics. - : Frontiers Media S.A.. - 2296-424X. ; 12
-
Forskningsöversikt (refereegranskat)abstract
- In the context of the 21st century and the fourth industrial revolution, the substantial proliferation of data has established it as a valuable resource, fostering enhanced computational capabilities across scientific disciplines, including physics. The integration of Machine Learning stands as a prominent solution to unravel the intricacies inherent to scientific data. While diverse machine learning algorithms find utility in various branches of physics, there exists a need for a systematic framework for the application of Machine Learning to the field. This review offers a comprehensive exploration of the fundamental principles and algorithms of Machine Learning, with a focus on their implementation within distinct domains of physics. The review delves into the contemporary trends of Machine Learning application in condensed matter physics, biophysics, astrophysics, material science, and addresses emerging challenges. The potential for Machine Learning to revolutionize the comprehension of intricate physical phenomena is underscored. Nevertheless, persisting challenges in the form of more efficient and precise algorithm development are acknowledged within this review.
|
|
| 4. |
- Sztuk-Dambietz, J., et al.
(författare)
-
Operational experience with Adaptive Gain Integrating Pixel Detectors at European XFEL
- 2024
-
Ingår i: Frontiers in Physics. - : Frontiers Media SA. - 2296-424X. ; 11
-
Forskningsöversikt (refereegranskat)abstract
- The European X-ray Free Electron Laser (European XFEL) is a cutting-edge user facility that generates per second up to 27,000 ultra-short, spatially coherent X-ray pulses within an energy range of 0.26 to more than 20 keV. Specialized instrumentation, including various 2D X-ray detectors capable of handling the unique time structure of the beam, is required. The one-megapixel AGIPD (AGIPD1M) detectors, developed for the European XFEL by the AGIPD Consortium, are the primary detectors used for user experiments at the SPB/SFX and MID instruments. The first AGIPD1M detector was installed at SPB/SFX when the facility began operation in 2017, and the second one was installed at MID in November 2018. The AGIPD detector systems require a dedicated infrastructure, well-defined safety systems, and high-level control procedures to ensure stable and safe operation. As of now, the AGIPD1M detectors installed at the SPB/SFX and MID experimental end stations are fully integrated into the European XFEL environment, including mechanical integration, vacuum, power, control, data acquisition, and data processing systems. Specific high-level procedures allow facilitated detector control, and dedicated interlock systems based on Programmable Logic Controllers ensure detector safety in case of power, vacuum, or cooling failure. The first 6 years of operation have clearly demonstrated that the AGIPD1M detectors provide high-quality scientific results. The collected data, along with additional dedicated studies, have also enabled the identification and quantification of issues related to detector performance, ensuring stable operation. Characterization and calibration of detectors are among the most critical and challenging aspects of operation due to their complex nature. A methodology has been developed to enable detector characterization and data correction, both in near real-time (online) and offline mode. The calibration process optimizes detector performance and ensures the highest quality of experimental results. Overall, the experience gained from integrating and operating the AGIPD detectors at the European XFEL, along with the developed methodology for detector characterization and calibration, provides valuable insights for the development of next-generation detectors for Free Electron Laser X-ray sources.
|
|
