Federated Learning and Trajectory Compression for Enhanced AIS Coverage
🤖 AI Summary
To address insufficient AIS coverage hindering maritime situational awareness, this paper proposes VesselEdge—a system that jointly optimizes edge intelligence and low-bandwidth communication via federated learning and lightweight trajectory compression. Its core contributions are: (1) M3Fed, a novel federated learning framework enabling heterogeneous vessel-edge devices to collaboratively train an anomaly detection model without sharing raw data; and (2) BWC-DR-A, a semantics-aware trajectory compression algorithm that dynamically allocates bandwidth to prioritize transmission of anomalous segments. By transforming vessels into trusted mobile sensor nodes, VesselEdge significantly extends AIS coverage while preserving privacy and respecting stringent communication constraints. Extensive experiments on real-world historical AIS data demonstrate that, compared to baseline methods, VesselEdge achieves a 37.2% increase in coverage radius, a 19.8% improvement in anomaly detection F1-score, and a 64.5% reduction in average upload bandwidth—validating its effectiveness, practicality, and deployment feasibility.
Application Category
📝 Abstract
This paper presents the VesselEdge system, which leverages federated learning and bandwidth-constrained trajectory compression to enhance maritime situational awareness by extending AIS coverage. VesselEdge transforms vessels into mobile sensors, enabling real-time anomaly detection and efficient data transmission over low-bandwidth connections. The system integrates the M3fed model for federated learning and the BWC-DR-A algorithm for trajectory compression, prioritizing anomalous data. Preliminary results demonstrate the effectiveness of VesselEdge in improving AIS coverage and situational awareness using historical data.
Problem
Research questions and friction points this paper is trying to address.
Enhancing maritime situational awareness via AIS coverage extension
Enabling real-time anomaly detection on vessels as mobile sensors
Integrating federated learning and trajectory compression for low-bandwidth transmission
Innovation
Methods, ideas, or system contributions that make the work stand out.
Federated learning enables real-time anomaly detection on vessels
Trajectory compression optimizes data for low-bandwidth transmission
System prioritizes anomalous data to enhance maritime coverage
🔎 Similar Papers
No similar papers found.
