PolymoRF: Polymorphic Wireless Receivers Through Physical-Layer Deep Learning

TL;DR

PolymoRF introduces a deep learning-based polymorphic wireless receiver capable of real-time reconfiguration by inferring waveform parameters, significantly improving flexibility and efficiency in wireless communications.

Contribution

The paper presents RFNet, a novel embedded deep learning architecture, and a hardware/software system that enable real-time waveform inference and reconfiguration in wireless receivers.

Findings

RFNet achieves accuracy comparable to state-of-the-art methods.

PolymoRF reduces latency and hardware complexity by 52x and 8x respectively.

System throughput reaches 87% of an ideal oracle system.

Abstract

Today's wireless technologies are largely based on inflexible designs, which makes them inefficient and prone to a variety of wireless attacks. To address this key issue, wireless receivers will need to (i) infer on-the-fly the physical-layer parameters currently used by transmitters; and if needed, (ii) change their hardware and software structures to demodulate the incoming waveform. In this paper, we introduce PolymoRF, a deep learning-based polymorphic receiver able to reconfigure itself in real time based on the inferred waveform parameters. Our key technical innovations are (i) a novel embedded deep learning architecture, called RFNet, which enables the solution of key waveform inference problems; (ii) a generalized hardware/software architecture that integrates RFNet with radio components and signal processing. We prototype PolymoRF on a custom software-defined radio platform, and show through extensive over-the-air experiments that (i) RFNet achieves similar accuracy to that of state-of-the-art yet with 52x and 8x latency and hardware reduction; (ii) PolymoRF achieves throughput within 87% of a perfect-knowledge Oracle system, thus demonstrating for the first time that polymorphic receivers are feasible and effective.