References & Citations
Computer Science > Networking and Internet Architecture
Title: Hidden Markov Model-Based Encoding for Time-Correlated IoT Sources
(Submitted on 19 Jan 2021 (v1), last revised 20 Jan 2021 (this version, v2))
Abstract: As the use of Internet of Things (IoT) devices for monitoring purposes becomes ubiquitous, the efficiency of sensor communication is a major issue for the modern Internet. Channel coding is less efficient for extremely short packets, and traditional techniques that rely on source compression require extensive signaling or pre-existing knowledge of the source dynamics. In this work, we propose an encoding and decoding scheme that learns source dynamics online using a Hidden Markov Model (HMM), puncturing a short packet code to outperform existing compression-based approaches. Our approach shows significant performance improvements for sources that are highly correlated in time, with no additional complexity on the sender side.
Submission history
From: Federico Chiariotti [view email][v1] Tue, 19 Jan 2021 09:45:27 GMT (2510kb,D)
[v2] Wed, 20 Jan 2021 11:45:48 GMT (342kb)
Link back to: arXiv, form interface, contact.