TinyML-Based Adaptive Pulse Shaping for Edge Intelligence in IoT/IIoT
Afan Ali
TL;DR
This paper presents a TinyML-based adaptive pulse shaping method for edge devices in IoT/IIoT, optimizing communication performance with a lightweight neural network that reduces PAPR and enhances energy efficiency.
Contribution
It introduces a novel TinyML neural network for adaptive pulse shaping, optimized for resource-constrained edge devices in IoT/IIoT communication systems.
Findings
Achieves up to 2 dB PAPR reduction compared to traditional filters
Validated through extensive simulations in DFT-s-OFDM systems
Offers an energy-efficient, scalable solution for IoT/IIoT applications
Abstract
Edge intelligence in IoT and IIoT demands lightweight algorithms for data processing on resource-constrained devices. This paper introduces a novel adaptive pulse shape filter based on TinyML for PAPR and SER optimization on edge devices used in uplink IoT communication. Implemented on IoT nodes such as sensors, our pruned neural network provides up to 2 dB PAPR saving over root-raised-cosine (RRC) filters. Mass simulations validate its efficacy in DFT-s-OFDM systems and offer an energy-efficient and scalable solution for IoT/IIoT use cases such as smart factories and rural connectivity.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.