Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

SALINA: Towards Sustainable Live Sonar Analytics in Wild Ecosystems

Chi Xu, Rongsheng Qian, Hao Fang, Xiaoqiang Ma, William I. Atlas, Jiangchuan Liu, Mark A. Spoljaric

Abstract

Sonar radar captures visual representations of underwater objects and structures using sound wave reflections, making it essential for exploration, mapping, and continuous surveillance in wild ecosystems. Real-time analysis of sonar data is crucial for time-sensitive applications, including environmental anomaly detection and in-season fishery management, where rapid decision-making is needed. However, the lack of both relevant datasets and pre-trained DNN models, coupled with resource limitations in wild environments, hinders the effective deployment and continuous operation of live sonar analytics. We present SALINA, a sustainable live sonar analytics system designed to address these challenges. SALINA enables real-time processing of acoustic sonar data with spatial and temporal adaptations, and features energy-efficient operation through a robust energy management module. Deployed for six months at two inland rivers in British Columbia, Canada, SALINA provided continuous 24/7 underwater monitoring, supporting fishery stewardship and wildlife restoration efforts. Through extensive real-world testing, SALINA demonstrated an up to 9.5% improvement in average precision and a 10.1% increase in tracking metrics. The energy management module successfully handled extreme weather, preventing outages and reducing contingency costs. These results offer valuable insights for long-term deployment of acoustic data systems in the wild.

SALINA: Towards Sustainable Live Sonar Analytics in Wild Ecosystems

Abstract

Sonar radar captures visual representations of underwater objects and structures using sound wave reflections, making it essential for exploration, mapping, and continuous surveillance in wild ecosystems. Real-time analysis of sonar data is crucial for time-sensitive applications, including environmental anomaly detection and in-season fishery management, where rapid decision-making is needed. However, the lack of both relevant datasets and pre-trained DNN models, coupled with resource limitations in wild environments, hinders the effective deployment and continuous operation of live sonar analytics. We present SALINA, a sustainable live sonar analytics system designed to address these challenges. SALINA enables real-time processing of acoustic sonar data with spatial and temporal adaptations, and features energy-efficient operation through a robust energy management module. Deployed for six months at two inland rivers in British Columbia, Canada, SALINA provided continuous 24/7 underwater monitoring, supporting fishery stewardship and wildlife restoration efforts. Through extensive real-world testing, SALINA demonstrated an up to 9.5% improvement in average precision and a 10.1% increase in tracking metrics. The energy management module successfully handled extreme weather, preventing outages and reducing contingency costs. These results offer valuable insights for long-term deployment of acoustic data systems in the wild.

Paper Structure

This paper contains 29 sections, 9 equations, 20 figures, 5 tables.

Table of Contents

  1. Introduction
  2. Background and Motivations
  3. Sonar Frame Capture
  4. Characteristics of Live Sonar Analytics
  5. Deployment Challenges
  6. SALINA Overview
  7. Data Preprocessing and Wrangling
  8. Sonar Dataset Preparation
  9. DNN Model Adaptations
  10. Sustainable Sonar Streaming
  11. Sonar Data Preprocessing and Wrangling
  12. Denoising and Background Removal
  13. Hedging Design in Channel Population
  14. DNN Model Adaptation
  15. On Premise Inference with CNN
  16. ...and 14 more sections

Figures (20)

  • Figure 1: (a) a mounted ARIS sonar radar, (b) sample frames from ARIS sonar.
  • Figure 2: Sonar radar and edge devices deployed in off-grid temperate forest.
  • Figure 3: Unique observations in sonar frames. The left image shows reverberation effects (elongated curves) and pervasive speckle noise. The right images display three consecutive frames (F1, F2, and F3) illustrating the moving acoustic shadow caused by a coho salmon. The ghost objects are shown as stacked highlighted areas above the fish.
  • Figure 4: Acoustic shadows in sonar frames.
  • Figure 5: Module design.
  • ...and 15 more figures