arXiv subjects

Alperen Aksoy

Publications and source records attributed to Alperen Aksoy.

3 recordsLinked to original sources

arXiv2026

A dual-polarization whitened-template trigger for real-time radio detection of extensive air showers

Autonomous radio stations require a first-stage trigger that rejects measured background while preserving weak extensive-air-shower pulses under tight hardware constraints. We present a dual-polarization trigger based on a 16-sample whitened pulse template evaluated continuously on the orthogonal north-south and east-west channels of an Auger Engineering Radio Array station. The template is derived from detector-folded air-shower pulses and whitened using the covariance of measured background. Responses from both polarizations are combined within a fixed timing radius and compared with a calibrated threshold. On an independent test set, the trigger reaches an efficiency of 0.97, with a 95 percent confidence interval of 0.96 to 0.98, at a frame-equivalent candidate rate of 3.39 kHz. At the same operating point, an unwhitened dual-polarization template reaches 0.41 efficiency and an amplitude trigger reaches 0.16. Efficiency remains at least 0.94 in every populated signal-to-noise bin. Replay of 159.29 ms of continuous measured background yields a score-cluster rate of 6.64 kHz, below the 71.4 kHz capacity of the downstream module. The FPGA implementation accepts one dual-polarization sample per clock, closes timing at 200 MHz, and uses 2377 lookup tables, 5330 registers, and 32 digital signal-processing blocks. Register-transfer-level simulation agrees bit for bit with the reference implementation for 400 traces. These results show that background covariance, pulse morphology, and dual-polarization consistency can be combined in a compact real-time radio trigger.

astro-ph.IM↗

arXiv2026

Hybrid neural denoising for resource-efficient near- and sub-threshold radio triggering of extensive air showers

Autonomous radio self-triggering for extensive air showers must reject variable radio-frequency interference while preserving sensitivity to weak pulses and remaining compatible with station-level edge hardware. This work presents a hybrid neural trigger in which waveform recovery and signal classification are treated as a single deployment-constrained problem. A compact convolutional denoiser maps a noisy single-channel trace to a cleaned estimate of the air-shower pulse, which is then evaluated by a compact classifier. The method is tested with measured high-interference background traces and detector-folded air-shower pulses from the Pierre Auger Offline simulation chain, with signals concentrated in the near- and sub-threshold regime. Model selection and deployment are linked through hyperparameter optimisation, quantisation-aware training, fixed-point quantisation, hls4ml firmware export, high-level synthesis, and register-transfer-level validation. The denoiser alone turns a simple peak-envelope decision into an efficient weak-pulse trigger, showing that the cleaned waveform carries trigger-relevant information beyond a final classifier score. In the full denoiser-classifier chain, the hybrid trigger improves signal-background separation and efficiency at fixed false-positive rates: at a false-positive rate of 10^-4, it retains about 41% of held-out signal traces in the weak-signal benchmark, while the classical peak-envelope trigger retains none. The cleaned waveform preserves timing and peak-amplitude structure for station-level diagnostics, feature extraction, and selective readout. The firmware meets timing on representative FPGA targets with microsecond-scale latency and compact arithmetic-resource demand. These results establish hybrid neural denoising as a practical route toward radio-only triggering for weak and inclined air-shower signals in noisy environments.

astro-ph.IM↗

arXiv2025

Quantization Effects of Artificial Neural Networks for Embedded Edge-Computing Applications

This paper examines the use of Quantized Neural Networks (QNNs) for two resource-constrained scientific applications: automated calibration of semi- conductor quantum bits (qubits) and scientific particle detectors. We evaluate the trade-offs between Post-Training Quantization (PTQ), Quantization-Aware Train- ing (QAT), and ultra-low-bit Binary Neural Networks (BNNs) with respect to la- tency and resource usage. Our results demonstrate that PTQ and QAT are easily implementable solutions achieving a four-fold reduction in memory usage for U- shaped CNN (U-Net) architectures, whereas BNNs are better suitable for use cases with challenging latency requirements. For the training of non-differentiable custom BNNs , we propose a novel, hardware-constrained learning approach using Genetic Algorithms (GAs). We showcase a Look-Up Table (LUT)-based BNN architecture suitable for direct conversion to Very High-Speed Integrated Circuit Hardware De- scription Language (VHDL) via the HCL4BNN framework. This method achieves nanosecond-scale inference latencies at 10 ns to 15 ns without requiring specialized Digital Signal Processor (DSP) or Block RAM (BRAM) resources.

cs.NE↗