Resource efficient data transmission on animals based on machine learning

2025-03-13

AI Summary

To address energy bottlenecks in wildlife tracking—specifically storage, computation, and radio transmission overheads imposed by stringent size and weight constraints on biologgers—this paper proposes a hardware-agnostic edge-intelligence data selection mechanism. The method deploys lightweight temporal models (e.g., TinyLSTM) directly on resource-constrained biologgers, enabling real-time feature extraction and adaptive threshold-based decision-making guided by behavioral semantics. This facilitates dynamic, on-device filtering of raw sensor data prior to transmission. Evaluated in real-world field deployments, the approach reduces radio transmission frequency substantially, yielding a 68% average reduction in transmission energy consumption and extending device battery lifetime by 2.3×, while preserving >92% accuracy in detecting critical behavioral events. The work establishes a scalable, embedded machine learning paradigm for low-power bio-telemetry, eliminating the need for hardware modifications or cloud dependency.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Computer Vision: Motion & Tracking
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Bio-loggers, electronic devices used to track animal behaviour through various sensors, have become essential in wildlife research. Despite continuous improvements in their capabilities, bio-loggers still face significant limitations in storage, processing, and data transmission due to the constraints of size and weight, which are necessary to avoid disturbing the animals. This study aims to explore how selective data transmission, guided by machine learning, can reduce the energy consumption of bio-loggers, thereby extending their operational lifespan without requiring hardware modifications.

Problem

• Reducing energy consumption in bio-loggers
• Extending operational lifespan without hardware changes
• Selective data transmission using machine learning

Innovation

• Machine learning optimizes data transmission.
• Selective transmission reduces energy consumption.
• Extends bio-logger lifespan without hardware changes.