arXiv subjects

T. Koga

Publications and source records attributed to T. Koga.

17 recordsLinked to original sources

arXiv2026

Commissioning and Low Latency Operation of the Graph Neural Network Electromagnetic Calorimeter Trigger at the Belle II Experiment

We present the commissioning and operation of the Graph Neural Network Electromagnetic Calorimeter Trigger Module (GNN-ETM) of the Belle II experiment at the SuperKEKB collider. The GNN-ETM processes calorimeter trigger cells as graph nodes to perform clustering and feature extraction. We fully integrate the system with the successive stages of the first-level trigger, develop slow-control drivers, and add online monitoring capabilities. We optimise the existing FPGA-based architecture through hardware-algorithm co-design, achieving an overall system latency of 1.053 us. Our hardware implementation is validated through register-transfer-level simulations, achieving bit-accurate agreement with the offline reference model. Online monitoring enables the measurement of instantaneous trigger rates, providing a quantitative basis for trigger-level performance studies. In summary, we report on the GNN-ETM as a fully operational, low-latency trigger module with online control and monitoring capabilities, compatible with the latency requirements of the Belle II first-level trigger system.

hep-ex↗

arXiv2026

Real-time graph neural networks on FPGAs for the Belle II electromagnetic calorimeter

We present the development and evaluation of a real-time Graph Neural Network-based trigger module for the electromagnetic calorimeter of the Belle~II experiment at the SuperKEKB collider. The algorithm processes calorimeter trigger cells as graph nodes to perform clustering, feature extraction, and per-cluster signal classification with deterministic latency. The model predicts cluster positions and energies and provides a signal classification score, enabling a more flexible clustering strategy than the baseline trigger algorithm. Implemented on an FPGA and integrated into the Belle~II trigger readout infrastructure for synchronous operation, the system sustains the MHz trigger throughput with an end-to-end latency of $3.168\,\mu$s. The performance is evaluated on simulated events and collision data. The energy resolution is comparable to the baseline trigger, while the position resolution for high-energy clusters improves by up to 18% in the central detector region. Cluster purity increases by up to 20% at low energies for isolated clusters, and cluster efficiency improves by up to 20% for overlapping clusters. The signal classifier enables additional background suppression at fixed signal retention. These results demonstrate a first step towards GNN-based real-time reconstruction on FPGAs in a collider trigger. While the end-to-end latency exceeds the trigger decision budget, the system already sustains full operational conditions with 100% uptime.

physics.ins-det↗

arXiv2025

Development of Deep Neural Network First-Level Hardware Track Trigger for the Belle II Experiment

The Belle II experiment at the SuperKEKB accelerator is designed to explore physics beyond the Standard Model with unprecedented luminosity. As the beam intensity increased, the experiment faced significant challenges due to higher beam-induced background, leading to a high trigger rate and placing limitations on further luminosity increases. To address this problem, we developed trigger logic for tracking using deep neural network (DNN) technology on an FPGA for the Belle II hardware trigger system, employing high-level synthesis techniques. By leveraging drift time and hit pattern information from the Central Drift Chamber and incorporating a simplified self-attention architecture, the DNN track trigger significantly improves track reconstruction performance at the hardware level. Compared to the existing neural track trigger, our implementation reduces the total track trigger rate by 37% while improving average efficiency for the signal tracks from 96% to 98% for charged tracks with transverse momentum > 0.3 GeV. This upgrade ensures the long-term viability of the Belle II data acquisition system as luminosity continues to increase.

physics.ins-det↗

arXiv2025

Design of the Global Reconstruction Logic in the Belle II Level-1 Trigger system

The Belle~II experiment is designed to search for physics beyond the Standard Model by investigating rare decays at the SuperKEKB $e^{+}e^{-}$ collider. Owing to the significant beam background at high luminosity, the data acquisition system employs a hardware-based Level-1~Trigger to reduce the readout data throughput by selecting collision events of interest in real time. The Belle~II Level-1~Trigger system utilizes FPGAs to reconstruct various detector observables from the raw data for trigger decision-making. The Global Reconstruction Logic receives these processed observables from four sub-trigger systems and provides a global summary for the final trigger decision. Its logic encompasses charged particle tracking, matching between sub-triggers, and the identification of special event topologies associated with low-multiplicity decays. This article discusses the hardware devices, FPGA firmware, integration with peripheral systems, and the design and performance of the trigger algorithms implemented within the Global Reconstruction Logic.

hep-ex↗

arXiv2024

The Belle II Detector Upgrades Framework Conceptual Design Report

We describe the planned near-term and potential longer-term upgrades of the Belle II detector at the SuperKEKB electron-positron collider operating at the KEK laboratory in Tsukuba, Japan. These upgrades will allow increasingly sensitive searches for possible new physics beyond the Standard Model in flavor, tau, electroweak and dark sector physics that are both complementary to and competitive with the LHC and other experiments.

hep-ex↗

arXiv2024

The Neural Network First-Level Hardware Track Trigger of the Belle II Experiment

We describe the principles and performance of the first-level ("L1") hardware track trigger of Belle II, based on neural networks. The networks use as input the results from the standard Belle II trigger, which provides "2D" track candidates in the plane transverse to the electron-positron beams. The networks then provide estimates for the origin of the 2D track candidates in direction of the colliding beams ("$z$-vertex"), as well as their polar emission angles $\theta$. Given the $z$-vertices of the "neural" tracks allows identifying events coming from the collision region ($z \approx 0$), and suppressing the overwhelming background from outside by a suitable cut $d$. Requiring $|z| < d$ for at least one neural track in an event with two or more 2D candidates will set an L1 trigger. The networks also enable a minimum bias trigger, requiring a single 2D track candidate validated by a neural track with a momentum larger than 0.7 GeV in addition to the $|z|$ condition. The momentum of the neural track is derived with the help of the polar angle $\theta$.

hep-ex↗

arXiv2020

The trigger slow control system of the Belle II experiment

The Belle II experiment at the SuperKEKB $e^{+}e^{-}$ collider in KEK, Japan, started physics data-taking with a complete detector from early 2019 with the primary physics goal of probing new physics in heavy quark and lepton decays. An online trigger system is indispensable for the Belle II experiment to reduce the beam background events associated with high electron and positron beam currents without sacrificing the target physics-oriented events. During the Belle II operation upon beam collision, the trigger system must be consistently controlled and its status must be carefully monitored in the process of data acquisition against unexpected situations. For this purpose, we have developed a slow control system for the Belle II trigger system. Around seventy thousand configuration parameters are saved in the Belle II central database server for every run when a run starts and stops. These parameters play an essential role in offline validation of the quality of runs. Around three thousand real-time variables are stored in the Belle II main archiving server, and the trend of some of these variables are regularly used for online and offline monitoring purposes. Various operator interface tools have been prepared and used. When the configuration parameters are not correctly applied, or some of the processes are unexpectedly terminated, the slow control system detects it, stops the data-taking process, and generates an alarm. In this article, we report how we constructed the Belle II trigger slow control system, and how we successfully managed to operate during its initial stage.

physics.ins-det↗

arXiv2020

First measurement of $\barν_μ$ and $ν_μ$ charged-current inclusive interactions on water using a nuclear emulsion detector

This paper reports the track multiplicity and kinematics of muons, charged pions, and protons from charged-current inclusive $\barν_μ$ and $ν_μ$ interactions on a water target, measured using a nuclear emulsion detector in the NINJA experiment. A 3-kg water target was exposed to the T2K antineutrino-enhanced beam with a mean energy of 1.3 GeV. Owing to the high-granularity of the nuclear emulsion, protons with momenta down to 200 MeV/$c$ from the neutrino-water interactions were detected. We find good agreement between the observed data and model predictions for all kinematic distributions other than the number of charged pions. These results demonstrate the capability of measurements with nuclear emulsion to improve neutrino interaction models.

hep-ex↗

arXiv2017

Baby MIND: A magnetized segmented neutrino detector for the WAGASCI experiment

T2K (Tokai-to-Kamioka) is a long-baseline neutrino experiment in Japan designed to study various parameters of neutrino oscillations. A near detector complex (ND280) is located 280~m downstream of the production target and measures neutrino beam parameters before any oscillations occur. ND280's measurements are used to predict the number and spectra of neutrinos in the Super-Kamiokande detector at the distance of 295~km. The difference in the target material between the far (water) and near (scintillator, hydrocarbon) detectors leads to the main non-cancelling systematic uncertainty for the oscillation analysis. In order to reduce this uncertainty a new WAter-Grid-And-SCintillator detector (WAGASCI) has been developed. A magnetized iron neutrino detector (Baby MIND) will be used to measure momentum and charge identification of the outgoing muons from charged current interactions. The Baby MIND modules are composed of magnetized iron plates and long plastic scintillator bars read out at the both ends with wavelength shifting fibers and silicon photomultipliers. The front-end electronics board has been developed to perform the readout and digitization of the signals from the scintillator bars. Detector elements were tested with cosmic rays and in the PS beam at CERN. The obtained results are presented in this paper.

physics.ins-det↗

arXiv2017

First neutrino event detection with nuclear emulsion at J-PARC neutrino beamline

Precise neutrino--nucleus interaction measurements in the sub-multi GeV region are important to reduce the systematic uncertainty in future neutrino oscillation experiments. Furthermore, the excess of ${ν_e}$ interactions, as a possible interpretation of the existence of a sterile neutrino has been observed in such an energy region. The nuclear emulsion technique can measure all the final state particles with low energy threshold for a variety of targets (Fe, C, H${_2}$O, and so on). Its sub-$μ$m position resolution allows measurements of the ${ν_e}$ cross-section with good electron/gamma separation capability. We started a new experiment at J-PARC to study sub-multi GeV neutrino interactions by introducing the nuclear emulsion technique. The J-PARC T60 experiment has been implemented as a first step of such a project. Systematic neutrino event analysis with full scanning data in the nuclear emulsion detector was performed for the first time. The first neutrino event detection and its analysis is described in this paper.

hep-ex↗

arXiv2017

Baby MIND Experiment Construction Status

Baby MIND is a magnetized iron neutrino detector, with novel design features, and is planned to serve as a downstream magnetized muon spectrometer for the WAGASCI experiment on the T2K neutrino beam line in Japan. One of the main goals of this experiment is to reduce systematic uncertainties relevant to CP-violation searches, by measuring the neutrino contamination in the anti-neutrino beam mode of T2K. Baby MIND is currently being constructed at CERN, and is planned to be operational in Japan in October 2017.

physics.ins-det↗

arXiv2017

Baby MIND: A magnetised spectrometer for the WAGASCI experiment

The WAGASCI experiment being built at the J-PARC neutrino beam line will measure the difference in cross sections from neutrinos interacting with a water and scintillator targets, in order to constrain neutrino cross sections, essential for the T2K neutrino oscillation measurements. A prototype Magnetised Iron Neutrino Detector (MIND), called Baby MIND, is being constructed at CERN to act as a magnetic spectrometer behind the main WAGASCI target to be able to measure the charge and momentum of the outgoing muon from neutrino charged current interactions.

physics.ins-det↗

arXiv2017

First demonstration of emulsion multi-stage shifter for accelerator neutrino experiment in J-PARC T60

We describe the first ever implementation of an emulsion multi-stage shifter in an accelerator neutrino experiment. The system was installed in the neutrino monitor building in J-PARC as a part of a test experiment T60 and stable operation was maintained for a total of 126.6 days. By applying time information to emulsion films, various results were obtained. Time resolutions of 5.3 to 14.7 s were evaluated in an operation spanning 46.9 days (time resolved numbers of 3.8--1.4$\times10^{5}$). By using timing and spatial information, a reconstruction of coincident events that consisted of high multiplicity events and vertex events, including neutrino events was performed. Emulsion events were matched to events observed by INGRID, one of near detectors of the T2K experiment, with high reliability (98.5\%) and hybrid analysis was established via use of the multi-stage shifter. The results demonstrate that the multi-stage shifter is feasible for use in neutrino experiments.

physics.ins-det↗

arXiv2013

Spin blocker made of semiconductor double quantum well using the Rashba effect

We propose a lateral spin-blockade device that uses an InGaAs/InAlAs double quantum well (DQW), where the values of the Rashba spin-orbit parameter $α_{\rm R}$ are opposite in sign but equal in magnitude between the constituent quantum wells (QW). By tuning the channel length of DQW and the magnitude of the externally applied in-plane magnetic field, one can block the transmission of one spin (e.g., spin-down) component, leading to a spin-polarized current. Such a spin-blocking effect, brought about by wave vector matching of the spin-split Fermi surfaces between the two QWs, paves the way for a new scheme of spin-polarized electric current generation for future spintronics applications based on semiconductor band engineering.

cond-mat.mes-hall↗

arXiv2004

Effect of an InP/In$_{0.53}$Ga$_{0.47}$As Interface on Spin-orbit Interaction in In$_{0.52}$Al$_{0.48}$As/In$_{0.53}$Ga$_{0.47}$As Heterostructures

We report the effect of the insertion of an InP/In$_{0.53}$Ga$_{47}$As Interface on Rashba spin-orbit interaction in In$_{0.52}$Al$_{0.48}$As/In$_{0.53}$Ga$_{0.47}$As quantum wells. A small spin split-off energy in InP produces a very intriguing band lineup in the valence bands in this system. With or without this InP layer above the In$_{0.53}$Ga$_{47}$As well, the overall values of the spin-orbit coupling constant $α$ turned out to be enhanced or diminished for samples with the front- or back-doping position, respectively. These experimental results, using weak antilocalization analysis, are compared with the results of the $\mathbf{k\cdot p}$ theory. The actual conditions of the interfaces and materials should account for the quantitative difference in magnitude between the measurements and calculations.

cond-mat.mes-hall↗

arXiv2004

Competition between Spin-Orbit Interaction and Zeeman Coupling in Rashba 2DEGs

We investigate systematically how the interplay between Rashba spin-orbit interaction and Zeeman coupling affects the electron transport and the spin dynamics in InGaAs-based 2D electron gases. From the quantitative analysis of the magnetoconductance, measured in the presence of an in-plane magnetic field, we conclude that this interplay results in a spin-induced breaking of time reversal symmetry and in an enhancement of the spin relaxation time. Both effects, due to a partial alignment of the electron spin along the applied magnetic field, are found to be in excellent agreement with recent theoretical predictions.

cond-mat.mes-hall↗

arXiv2003

Statistical significance of fine structure in the frequency spectrum of Aharonov-Bohm conductance oscillations

We discuss a statistical analysis of Aharonov-Bohm conductance oscillations measured in a two-dimensional ring, in the presence of Rashba spin-orbit interaction. Measurements performed at different values of gate voltage are used to calculate the ensemble-averaged modulus of the Fourier spectrum and, at each frequency, the standard deviation associated to the average. This allows us to prove the statistical significance of a splitting that we observe in the h/e peak of the averaged spectrum. Our work illustrates in detail the role of sample specific effects on the frequency spectrum of Aharonov-Bohm conductance oscillations and it demonstrates how fine structures of a different physical origin can be discriminated from sample specific features.

cond-mat.mes-hall↗