Summary

International Conference on Emerging Technologies for Communications

2020

Session Number:P1

Session:

Number:P1-2

Real-Time Optimization of Compression Ratio in Distributed Compressed Sensing based on Multi-Agent Deep Reinforcement Learning for Edge Computing

Masatoshi Sekine,  Satoshi Ikada,  

pp.-

Publication Date:2020/12/2

Online ISSN:2188-5079

DOI:10.34385/proc.63.P1-2

PDF download

PayPerView

Summary:
In this paper, we propose a lightweight and adaptive distributed compressed sensing (DCS) with multi-sensor collaboration based on multi-agent deep reinforcement learning (LADICS-MARL).It is important to efficiently acquire data generated by sensor nodes deployed over a wide area and for long periods in monitoring systems for social infrastructures and factory equipment. We previously proposed a lightweight and adaptive compressed sensing method based on deep learning for edge devices, called LACSLE, that changes the compression ratio in real-time according to the raw data. Our new proposed method is an extension for multiple senders and one receiver and supports DCS through which multiple compressed data are simultaneously reconstructed. A performance evaluation using acceleration data from multiple sensor terminals acquired on a bridge suggests that the multi-agent-based LADICS-MARL can reconstruct with less error and less data compared with the single-agent-based LACSLE.