RNO-G Beamforming Cuts Radio-Neutrino Trigger Thresholds

Delay-and-sum beams with power integration reduce the SNR needed for RNO-G triggers by 25% on average.

Editorial Desk·July 28, 2026·4 min readstrong

Underlying Paper

The beamformed trigger of RNO-G: its design and in-field performance

G CollaborationS. Agarwal (University of Kansas)J. A. Aguilar (Universit\'e Libre de Bruxelles)N. Alden (University of Chicago)S. Ali (University of Kansas)P. Allison (Ohio State University)M. Betts (Pennsylvania State UniversityDept. of PhysicsPennsylvania State UniversityDept. of Astronomy and Astrophysics)D. Besson (University of Kansas)A. Bishop (University of Wisconsin-Madison)O. Botner (Uppsala University)S. Bouma (Erlangen Centre for Astroparticle Physics)S. Buitink (Vrije Universiteit BrusselAstrophysical InstituteRadboud University)R. Camphyn (Universit\'e Libre de Bruxelles)J. Chan (University of Wisconsin-Madison)S. Chiche (Universit\'e Libre de Bruxelles)B. A. Clark (University of Maryland)K. Couberly (University of Kansas)D. Dakroub (University of Kansas)K. D. de Vries (Vrije Universiteit BrusselDienst ELEM)C. Deaconu (University of Chicago)P. Giri (University of Nebraska-Lincoln)C. Glaser (Uppsala UniversityTU Dortmund University)H. Gui (Ohio State University)A. Hallgren (Uppsala University)J. C. Hanson (Whittier College)S. Hassiki (University of Nebraska-Lincoln)K. Helbing (University of Wuppertal)B. Hendricks (Pennsylvania State UniversityDept. of PhysicsPennsylvania State UniversityCenter for Multimessenger Astrophysics)J. Henrichs (Deutsches Elektronen-Synchrotron DESYErlangen Centre for Astroparticle Physics)N. Heyer (Uppsala University)C. Hornhuber (University of Kansas)E. Huesca Santiago (Deutsches Elektronen-Synchrotron DESY)K. Hughes (Ohio State University)A. Jaitly (Deutsches Elektronen-Synchrotron DESYErlangen Centre for Astroparticle Physics)A. Karle (University of Wisconsin-Madison)J. L. Kelley (University of Wisconsin-Madison)C. Kopper (Erlangen Centre for Astroparticle Physics)M. Korntheuer (Universit\'e Libre de BruxellesVrije Universiteit BrusselDienst ELEM)M. Kowalski (Deutsches Elektronen-Synchrotron DESYHumboldt-Universit\"at zu Berlin)I. Kravchenko (University of Nebraska-Lincoln)R. Krebs (Pennsylvania State UniversityDept. of PhysicsPennsylvania State UniversityCenter for Multimessenger Astrophysics)M. Kugelmeier (University of Wisconsin-Madison)D. Kullgren (Pennsylvania State UniversityDept. of PhysicsPennsylvania State UniversityCenter for Multimessenger Astrophysics)R. Lahmann (Erlangen Centre for Astroparticle Physics)C. -H. Liu (University of Nebraska-Lincoln)Y. Liu (Ohio State University)M. J. Marsee (University of Alabama)K. Mulrey (Radboud University)M. Muzio (University of Wisconsin-Madison)A. Nelles (Deutsches Elektronen-Synchrotron DESYErlangen Centre for Astroparticle Physics)A. Novikov (University of Delaware)A. Nozdrina (Ohio State University)E. Oberla (University of Chicago)N. Punsuebsay (University of Delaware)M. Ravn (Uppsala University)Z. Riesen (Ohio State University)A. Rifaie (University of Wuppertal)D. Ryckbosch (Ghent University)F. Schl\"uter (Universit\'e Libre de Bruxelles)O. Scholten (Vrije Universiteit BrusselDienst ELEMUniversity of GroningenKapteyn Institute)P. Schriefer (Erlangen Centre for Astroparticle Physics)D. Seckel (University of Delaware)M. F. H. Seikh (University of Kansas)Z. S. Selcuk (Deutsches Elektronen-Synchrotron DESYErlangen Centre for Astroparticle Physics)J. Stachurska (Ghent University)J. Stoffels (Vrije Universiteit BrusselDienst ELEM)S. Toscano (Universit\'e Libre de Bruxelles)D. Tosi (University of Wisconsin-Madison)J. Tutt (Pennsylvania State UniversityDept. of Astronomy and Astrophysics)N. van Eijndhoven (Vrije Universiteit BrusselDienst ELEM)A. G. Vieregg (University of Chicago)A. Vijai (University of Maryland)H. Warnhofer (Deutsches Elektronen-Synchrotron DESYErlangen Centre for Astroparticle Physics)D. Washington (Pennsylvania State UniversityDept. of PhysicsPennsylvania State UniversityDept. of Astronomy and Astrophysics)C. Welling (Pennsylvania State UniversityDept. of PhysicsPennsylvania State UniversityDept. of Astronomy and AstrophysicsPennsylvania State UniversityCenter for Multimessenger Astrophysics)D. R. Williams (University of Alabama)P. Windischhofer (University of ChicagoDeutsches Elektronen-Synchrotron DESY)S. Wissel (Pennsylvania State UniversityDept. of PhysicsPennsylvania State UniversityDept. of Astronomy and AstrophysicsPennsylvania State UniversityCenter for Multimessenger Astrophysics)A. Zink (Erlangen Centre for Astroparticle Physics)

The Radio Neutrino Observatory in Greenland (RNO-G) is a neutrino detector under construction at Summit Station, with 8 out of a planned 35 stations currently deployed. We have designed and deployed a new phased array (PA) trigger based on delay-and-sum beamforming and power integration. This trigger improves detector performance by suppressing thermal noise and better targeting neutrino-induced Askaryan signals. The new PA trigger has been deployed since the 2025 season. The trigger performance has been characterized using test pulses and calibration pulsers both in the lab and in-situ, and we find across these tests a 25% average reduction in the signal-to-noise ratio (SNR) needed to trigger on signals. Simulations are shown to be representative of the detector, and simulated trigger efficiencies are within 10% of measured data. Following the in-situ trigger validation, we use data-driven trigger performance to inform simulations of the detector effective volumes. The PA trigger increases our effective volume significantly, by over a factor of 2 below 0.1 EeV and at least a factor of 1.3 at the highest energies of 100 EeV.

arXiv:2607.24470Submitted: Jul 28, 2026v1

Radio-neutrino observatories live or die by their trigger thresholds. Askaryan pulses from ultra-high-energy neutrinos are brief, broadband radio transients buried under thermal noise, so a trigger that rejects noise without rejecting real pulses directly changes the detector’s exposure. This paper reports the design and field performance of the new RNO-G phased-array trigger, deployed from the 2025 season, for an array that currently has 8 of its planned 35 Greenland stations in the ice.

Core Contribution

The central contribution is a practical trigger architecture for an operating neutrino detector, not a simulation-only optimization. The authors replace the older high-low threshold trigger as the primary deep trigger with a phased-array design: antenna waveforms are delayed, summed into beams, squared or power-integrated, and tested against thresholds. The point is to use coherence across the vertically spaced antennas to make Askaryan-like plane-wave impulses grow faster than incoherent thermal fluctuations.

Figure 1 gives the cleanest visual intuition. In an in-situ pulser event, the beam closest to the incoming RF direction has the largest coherent signal after upsampling and beamforming, while off-direction beams show weaker or destructively interfering traces.

Figure 1. The main testing method for trigger performance is through in-situ pulsers. This shows an example pulser event recorded in the FLOWER and each individual beam trace after upsampling and beamforming is performed. The beam direction aligning nearest to the received direction of the RF signal has the largest coherence and shows the largest signals. The most vertical beams (beams 0, 1, 10, 11) show distinct pulses from each channel with no interference. The center beams (beams 5, 6) show the maximum signal where the signals coherently align. The intermediary beams (beams 2, 3, 4, 7, 8, 9) show limited signals where they interfere destructively.

Technical Approach

The trigger pipeline is built around finite-rate digitized waveforms, digital upsampling, delay-and-sum beamforming, and a power-integration decision. The paper devotes substantial space to the firmware-facing details because they matter: interpolation changes pulse timing, the finite impulse response low-pass filter sets the usable bandwidth after zero-stuffing, and the integration window must be long enough to collect Askaryan pulse power without admitting too much thermal noise.

The authors compare several trigger variants in simulation, including idealized references, a Hilbert-envelope option, simple amplitude thresholds on beamformed traces, and power-averaging settings. Their selected design is the phased-array power-integration trigger with a 12.7 ns integration window. That choice is motivated by simulated Askaryan pulses across view angles from the Cherenkov angle: wider off-cone signals need longer windows, but too long a window starts to dilute the threshold advantage. Figure 6 summarizes this design trade-off against the high-low trigger and alternative beamformed triggers.

Figure 6. Comparison of various trigger methods and settings to the chosen PA trigger, ``PA PI. w/ 12.7ns Win.'', and the hi-lo trigger, ``HL: 4.0''. Ideal triggers are shown as a comparison to realized methods. Top left: The chosen trigger method in sec:design with different integration windows. Top right: The integration window in sec:design with different settings needed for a realistic implementation. Bottom left: An alternative trigger method using a Hilbert envelope. Bottom right: An alternative trigger method using a simple amplitude threshold on the beamformed voltage traces.

The validation path is also concrete. The paper uses laboratory tests, field pulsers, and embedded calibration sources to measure trigger efficiency as a function of signal-to-noise ratio. The SNR definition is tied to observed waveform amplitudes, with special handling for saturation at high amplitudes and thermal-noise saturation near low SNR. Figure 8 shows the measured efficiency procedure for one pulser and station: the attenuation scan maps pulser setting to waveform SNR, then an efficiency curve estimates the trigger turn-on.

Figure 8. Example efficiency measurement using embedded calibration sources. These are made for each station and pulser. Left: Mean SNR of the as-recorded waveforms plotted against the attenuation factor setting for pulser 0 on station 14. At SNRs around 3 the event SNR saturates to the SNR of thermal noise. At SNRs above 10, the signal amplitudes begin to saturate at the maximum voltages allowed by the RFoF diodes. Right: Efficiency curve for pulser 0 on station 14. The reported SNR in the efficiency curve is the second highest SNR from the attenuation fit to match studies in Ref.~RNO-G:2025inst, where the channel with the second highest SNR drives the 2 of 4 channel coincidence in the hi-lo trigger.

Results and Analysis

The headline result is a 25% average reduction in the SNR required to trigger, measured across the paper’s test configurations. For an instrument whose science reach is exposure-limited at the highest neutrino energies and threshold-limited at lower energies, that is a material gain: a lower trigger threshold admits fainter Askaryan pulses and therefore increases the volume of ice from which detectable signals can arrive.

The agreement between simulation and data is a second important result. The paper reports that simulated trigger efficiencies are within 10% of measured data after in-situ validation. That does not make the effective-volume estimate a direct measurement of neutrino sensitivity, but it does make the simulation chain more credible than an unconstrained design study. Using the data-informed trigger model, the authors estimate that the phased-array trigger increases RNO-G effective volume by more than a factor of 2 below 0.1 EeV and by at least 1.3× even at 100 EeV.

Those gains have different meanings across energy. Below 0.1 EeV, the factor-of-two improvement is the main physics argument for the trigger: threshold reduction changes whether weak signals enter the sample at all. At 100 EeV, where signals are stronger, the smaller but still positive 1.3× gain indicates that beamforming is not only a low-energy patch; it expands acceptance even when events are easier to trigger.

Caveats in Practice

The evidence is strong for trigger behavior and credible for effective-volume impact, but the scope is still bounded. RNO-G is not yet the full 35-station detector, and the validation uses pulser signals as controlled proxies for neutrino-induced Askaryan emission. The paper does compare pulser models with simulated Askaryan pulse structure, including frequency content and view-angle dependence, yet calibration pulsers cannot cover every geometry, ice-propagation condition, or hardware state expected in future data. The result is best read as a well-tested detector upgrade with a data-anchored sensitivity projection, rather than a completed measurement of neutrino performance from the full observatory.

Evidence Box

strong

Key Claims

  • Phased-array beamforming suppresses thermal-noise triggers
  • Power integration better targets Askaryan-like radio pulses
  • Data-informed trigger modeling increases projected RNO-G exposure
  • Simulation reproduces in-situ trigger efficiency closely enough for effective-volume studies

Key Results

  • 25% average reduction in SNR needed to trigger across lab and in-situ tests
  • Simulated trigger efficiencies within 10% of measured data
  • Effective volume increases by over 2× below 0.1 EeV
  • Effective volume increases by at least 1.3× at 100 EeV

Limitations & Caveats

  • Validation performed with calibration and test pulsers rather than detected neutrino events
  • Detector is partially deployed with 8 of 35 planned stations operating
  • Effective-volume gains depend on data-driven simulations after trigger validation
  • Pulser geometries and hardware states do not span all future in-ice signal conditions

Related Articles

Readers are encouraged to consult the original arXiv paper for complete details. SOTA Papers does not make claims beyond what is supported by the authors' reported evidence.