# Underwater Acoustic Channel Repository: Measured Channel Impulse Responses from Eight At-Sea Experiments

## Facts

- Provider: Zhengnan Li, Mandar Chitre, Diego Cuji, James Preisig, Andrew Singer, Milica Stojanovic, Paul van Walree
- License: CC-BY-4.0 (https://zenodo.org/records/21287414)
- Size: 73 files
- Format: CSV (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/UWA-CP/Underwater-Acoustic-Channel-Repository
- Access: public, readable without an account
- DOI: https://doi.org/10.5281/zenodo.21287414
- Category: Physics
- Source: https://doi.org/10.5281/zenodo.21287414
- Page: https://scigantic.com/archive/577274a1-a48e-5f2e-a63a-038b038b5874
- Keywords: Physics, CSV, Hugging Face

## About

Measured underwater acoustic channel impulse responses (CIRs) from at-sea experiments, collected for research on underwater acoustic communication, channel modeling and channel prediction. Eight channel groups, labeled Blue, Red, Yellow, Purple, Green, Black, Pink and Brown, span the North Atlantic, Singapore, Hawaii, Norway, Japan, the Mariana Trench and the Pacific Ocean, with ranges from tens of meters to thousands of kilometers, center frequencies from 75 Hz to 25 kHz, and receivers from single hydrophones to vertical, cross and planar arrays. Each channel is a compressed complex-baseband CIR matrix in MATLAB .mat format with a phase vector that stores the time-varying delay drift removed during compression, plus accompanying noise recordings. The Hugging Face copy keeps the original 67 files and names (about 22 GiB) in one folder per group under data/, with data/index.csv giving subset, role, size, MD5 and source DOI for every file. Load with scipy.io.loadmat; reconstruction tools in MATLAB and Python are maintained by the UWA Channels project. Published on Zenodo by Li, Chitre, Cuji, Preisig, Singer, Stojanovic and van Walree under CC-BY-4.0; cite doi:10.5281/zenodo.21287414.

## Load the data

```python
import pandas as pd
df = pd.read_csv("hf://datasets/UWA-CP/Underwater-Acoustic-Channel-Repository/data/index.csv")
df.head()
```

## Data structure

File types in sampled files: CSV

Example file: `data/index.csv`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/577274a1-a48e-5f2e-a63a-038b038b5874
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/577274a1-a48e-5f2e-a63a-038b038b5874
- Over MCP: `get_schema_card` with id `577274a1-a48e-5f2e-a63a-038b038b5874` at https://api.scigantic.com/mcp

## Related topics

- [CSV datasets](https://scigantic.com/topics/csv)
- [Physics datasets](https://scigantic.com/topics/physics)
