Skip to content

OpenWaveLogger v2026 (OWL-v2026): an open source, low cost, easy to build, high performance logger for wave data measurements

Home / Papers / OpenWaveLogger v2026 (OWL-v2026): an open source, low cost, easy to build, high performance logger for wave data measurements

OpenWaveLogger v2026 (OWL-v2026): an open source, low cost, easy to build, high performance logger for wave data measurements

Authors: Jean Rabault, Joey Voermans, Takuji Waseda, Takehiko Nose, Tsubasa Kodaira, Koya Sato, ALEXANDER BABANIN, Gaute Hope, Malte Müller, Lars Willas Dreyer, Øystein Lande, Atle Jensen, Øyvind Breivik Date: 2026-04-22 Paper ID: openalex:2604.20502

Summary

The OpenWaveLogger (OWL-v2026) is an open-source, low-cost platform designed for high-performance in-situ ocean wave measurements. By leveraging off-the-shelf components, the system enables researchers to collect full high-frequency time series (IMU and GNSS data) that are often inaccessible via telemetry-limited buoy networks. It achieves high-precision absolute UTC timestamping through GNSS pulse-per-second synchronization and offers reliable long-term battery autonomy for field deployments.

Key Contributions

  • Introduces the OpenWaveLogger (OWL-v2026), a low-cost (~$220) open-source hardware solution for high-frequency in-situ wave data acquisition.
  • Provides a custom firmware implementation enabling low-jitter, six-axis IMU logging at up to 416Hz and GNSS-synchronized timestamping with sub-10ms accuracy.
  • Demonstrates sustained 10-day continuous operation at 208Hz with 20-day autonomy on standard D-cell battery configurations.

Open Questions & Future Work

Archivist Review

This paper presents a hardware-focused contribution rather than an ML methodology. While the technical challenge of high-frequency data logging is relevant to time-series acquisition, the hardware design itself is not a reusable ML concept or framework. The open question regarding write latency is approved as it represents a fundamental physical bottleneck in long-term, high-frequency time-series data collection.

Approved Open Questions

  • Embedded High-Frequency Write Bottleneck: This bottleneck limits the scaling of sampling rates for long-term unattended field deployments where memory resources are constrained and data integrity is paramount.

Rejected Candidates

  • [open_question] SD card write latency bottlenecks (sd-card-write-latency-bottleneck) - other: Renamed to be more descriptive of the broader embedded constraint.

Metadata & Links

url
https://arxiv.org/abs/2604.20502
paper_id
2604.20502
paper_source
openalex
domain
time-series
tags
architectures
datasets
concept_slugs
dataset_slugs
skill
TimeSeriesSkill
created_at
2026-04-25T05:38:12Z