Explore the vast range of titles available.

Automated systems capable of real-time operation with minimal energy consumption are increasingly important in modern radio telescopes. Spiking Neural Networks (SNNs) promise efficient and dynamic spatio-temporal data processing. This paper reformulates a significant challenge in radio astronomy, Radio Frequency Interference (RFI) detection, as a time-series segmentation task suited for SNN execution. We explore several spectrogram encoding methods and network parameters, applying first and second-order leaky integrate and fire SNNs to tackle RFI detection. We introduce a divisive normalisation-inspired pre-processing step, improving detection performance across multiple encodings strategies. Our approach achieves competitive performance on a synthetic dataset and compelling initial results on real data from the Low-Frequency Array (LOFAR) establishing a baseline for future work. We position SNNs as a viable path towards real-time RFI detection, with many possibilities for follow-up studies. These findings highlight the potential for SNNs performing complex time-series tasks, paving the way towards efficient, real-time processing in radio astronomy and other data-intensive fields. This work addresses the challenges of radio frequency interference (RFI) in radio astronomy. The authors train spiking neural networks on synthetic and real data, demonstrating a viable path for real-time, energy-efficient RFI detection.

Imminent radio telescope observatories provide massive data rates making deep learning based processing appealing while simultaneously demanding real-time performance at low-energy; prohibiting the use of many artificial neural network based approaches. We begin tackling the scientifically existential challenge of Radio Frequency Interference (RFI) detection by deploying deep Spiking Neural Networks (SNNs) on resource-constrained neuromorphic hardware. Our approach partitions large, pre-trained networks onto SynSense Xylo hardware using maximal splitting, a novel greedy algorithm. We validate this pipeline with on-chip power measurements, achieving instrument-scaled inference at 100mW. While our full-scale SNN achieves state-of-the-art accuracy among SNN baselines, our experiments reveal a more important insight that a smaller un-partitioned model significantly outperforms larger, split models. This finding highlights that hardware co-design is paramount for optimal performance. Our work thus provides a practical deployment blueprint, a key insight into the challenges of model scaling, and reinforces radio astronomy as a demanding yet ideal domain for advancing applied neuromorphic computing.

Spiking Neural Networks (SNNs) promise efficient and dynamic spatio-temporal data processing. This paper reformulates a significant challenge in radio astronomy, Radio Frequency Interference (RFI) detection, as a time-series segmentation task suited for SNN execution. Automated systems capable of real-time operation with minimal energy consumption are increasingly important in modern radio telescopes. We explore several spectrogram encoding methods and network parameters, applying first and second-order leaky integrate and fire SNNs to tackle RFI detection. We introduce a divisive normalisation-inspired pre-processing step, improving detection performance across multiple encodings strategies. Our approach achieves competitive performance on a synthetic dataset and compelling initial results on real data from the Low-Frequency Array (LOFAR). We position SNNs as a viable path towards real-time RFI detection, with many possibilities for follow-up studies. These findings highlight the potential for SNNs performing complex time-series tasks, paving the way towards efficient, real-time processing in radio astronomy and other data-intensive fields.

Radio Frequency Interference (RFI) is a known growing challenge for radio astronomy, intensified by increasing observatory sensitivity and prevalence of orbital RFI sources. Spiking Neural Networks (SNNs) offer a promising solution for real-time RFI detection by exploiting the time-varying nature of radio observation and neuron dynamics together. This work explores the inclusion of polarisation information in SNN-based RFI detection, using simulated data from the Hydrogen Epoch of Reionisation Array (HERA) instrument and provides power usage estimates for deploying SNN-based RFI detection on existing neuromorphic hardware. Preliminary results demonstrate state-of-the-art detection accuracy and highlight possible extensive energy-efficiency gains.

Radio Frequency Interference (RFI) from anthropogenic radio sources poses significant challenges to current and future radio telescopes. Contemporary approaches to detecting RFI treat the task as a semantic segmentation problem on radio telescope spectrograms. Typically, complex heuristic algorithms handle this task of `flagging' in combination with manual labeling (in the most difficult cases). While recent machine-learning approaches have demonstrated high accuracy, they often fail to meet the stringent operational requirements of modern radio observatories. Owing to their inherently time-varying nature, spiking neural networks (SNNs) are a promising alternative method to RFI-detection by utilizing the time-varying nature of the spectrographic source data. In this work, we apply Liquid State Machines (LSMs), a class of spiking neural networks, to RFI-detection. We employ second-order Leaky Integrate-and-Fire (LiF) neurons, marking the first use of this architecture and neuron type for RFI-detection. We test three encoding methods and three increasingly complex readout layers, including a transformer decoder head, providing a hybrid of SNN and ANN techniques. Our methods extend LSMs beyond conventional classification tasks to fine-grained spatio-temporal segmentation. We train LSMs on simulated data derived from the Hyrogen Epoch of Reionization Array (HERA), a known benchmark for RFI-detection. Our model achieves a per-pixel accuracy of 98% and an F1-score of 0.743, demonstrating competitive performance on this highly challenging task. This work expands the sophistication of SNN techniques and architectures applied to RFI-detection, and highlights the effectiveness of LSMs in handling fine-grained, complex, spatio-temporal signal-processing tasks.

Gravitational waves are now routinely detected from compact binary mergers, with binary neutron star mergers being of note for multi-messenger astronomy as they have been observed to produce electromagnetic counterparts. Novel search pipelines for these mergers could increase the combined search sensitivity, and could improve the ability to detect real gravitational wave signals in the presence of glitches and non-stationary detector noise. Deep learning has found success in other areas of gravitational wave data analysis, but a sensitive deep learning-based search for binary neutron star mergers has proven elusive due to their long signal length. In this work, we present a deep learning pipeline for detecting binary neutron star mergers. By training a convolutional neural network to detect binary neutron star mergers in the signal-to-noise ratio time series, we concentrate signal power into a shorter and more consistent timescale than strain-based methods, while also being able to train our network to be robust against glitches. We compare our pipeline's sensitivity to the three offline detection pipelines using injections in real gravitational wave data, and find that our pipeline has a comparable sensitivity to the current pipelines below the 1 per 2 months detection threshold. Furthermore, we find that our pipeline can increase the total number of binary neutron star detections by 12% at a false alarm rate of 1 per 2 months. The pipeline is also able to successfully detect the two binary neutron star mergers detected so far by the LIGO-Virgo-KAGRA collaboration, GW170817 and GW190425, despite the loud glitch present in GW170817.

Gravitational wave detection has opened up new avenues for exploring and understanding some of the fundamental principles of the universe. The optimal method for detecting modelled gravitational-wave events involves template-based matched filtering and performing a multi-detector coincidence search in the resulting signal-to-noise ratio time series. In recent years, advancements in machine learning and deep learning have led to a flurry of research into using these techniques to replace matched filtering searches and for efficient and robust parameter estimation of the gravitational wave sources. This paper presents a feasibility study for a novel approach to detecting binary black hole gravitational wave signals, which utilizes deep learning techniques on the signal-to-noise ratio time series produced from matched filtering. We show that a deep-learning search can efficiently detect binary black hole gravitational waves from the signal-to-noise ratio time series in simulated Gaussian noise with simulated transient glitches. Furthermore, our search method can outperform a maximum SNR-based matched filtering search on simulated data of the Hanford and Livingston LIGO detectors in the presence of glitches. We further demonstrate that our approach can improve the detection sensitivity for binary black hole mergers at lower masses, relative to a baseline sensitivity of existing search pipelines and deep learning approaches. Lastly, since we are building upon the foundations of a matched filtering search pipeline, we can extract estimates for the signal-to-noise ratio and detector frame chirp mass of a gravitational wave event with similar accuracy as existing pipelines.

Radio Frequency Interference (RFI) poses a significant challenge in radio astronomy, arising from terrestrial and celestial sources, disrupting observations conducted by radio telescopes. Addressing RFI involves intricate heuristic algorithms, manual examination, and, increasingly, machine learning methods. Given the dynamic and temporal nature of radio astronomy observations, Spiking Neural Networks (SNNs) emerge as a promising approach. In this study, we cast RFI detection as a supervised multi-variate time-series segmentation problem. Notably, our investigation explores the encoding of radio astronomy visibility data for SNN inference, considering six encoding schemes: rate, latency, delta-modulation, and three variations of the step-forward algorithm. We train a small twolayer fully connected SNN on simulated data derived from the Hydrogen Epoch of Reionization Array (HERA) telescope and perform extensive hyper-parameter optimization. Results reveal that latency encoding exhibits superior performance, achieving a per-pixel accuracy of 98.8% and an f1-score of 0.761. Remarkably, these metrics approach those of contemporary RFI detection algorithms, notwithstanding the simplicity and compactness of our proposed network architecture. This study underscores the potential of RFI detection as a benchmark problem for SNN researchers, emphasizing the efficacy of SNNs in addressing complex time-series segmentation tasks in radio astronomy.

Computational workflow management systems power contemporary data-intensive sciences. The slowly resolving reproducibility crisis presents both a sobering warning and an opportunity to iterate on what science and data processing entails. The Square Kilometre Array (SKA), the world's largest radio telescope, is among the most extensive scientific projects underway and presents grand scientific collaboration and data-processing challenges. In this work, we aim to improve the ability of workflow management systems to facilitate reproducible, high-quality science. This work presents a scale and system-agnostic computational workflow model and extends five well-known reproducibility concepts into seven well-defined tenets for this workflow model. Additionally, we present a method to construct workflow execution signatures using cryptographic primitives in amortized constant time. We combine these three concepts and provide a concrete implementation in Data Activated Flow Graph Engine (DALiuGE), a workflow management system for the SKA to embed specific provenance information into workflow signatures, demonstrating the possibility of facilitating automatic formal verification of scientific quality in amortized constant time. We validate our approach with a simple yet representative astronomical processing task: filtering a noisy signal with a lowpass filter using CPU and GPU methods. This example shows the practicality and efficacy of combining formal tenet definitions with a workflow signature generation mechanism. Our framework, spanning formal UML specification, principled provenance information collection based on reproducibility tenets, and finally, a concrete example implementation in DALiuGE illuminates otherwise obscure scientific discrepancies and similarities between principally identical workflow executions.