Back to AI News
arxiv

Unified RF Source Separator Matches Specialized Models Without Knowing the Interference Type

Researchers show that specialized RF separators trained for a single signal/interference pair fail on unseen interference, often leaving the recovered signal worse than no processing at all. They instead train one interference-agnostic separator across all pairs using two architectures with no shared components: a convolutional WaveNet from the ICASSP 2024 RF Challenge and a band-split transformer adapted from music source separation. At the size of a single specialized model, the unified separator keeps bit error rate close to the models it replaces on nearly every pair and matches the best specialized model on interference absent from every training set, removing the need to know or label the interference type at test time.

AI Brief

Demonstrates a concrete, transferable research result — a single AI model for co-channel RF separation that generalizes to unseen jamming — valuable for readers building robust wireless and signal-processing systems.

arXiv:2611.17950v1 Announce Type: new Abstract: Separating a radio-frequency signal of interest from a co-channel interference is mostly addressed with specialized separators, each trained for one pair of signal and interference types and therefore relying on a label that a heavily jammed receiver cannot identify. We show that this reliance is not benign: applied to an interference it was not trained on, such a separator usually fails to recover the signal, and in most of the cases we measure it leaves the signal worse than no processing at all. We then train a single separator over every pair, giving it no indication at test time of which one it faces, on two architectures with no component in common: the convolutional WaveNet of the ICASSP 2024 RF Challenge, and a band-split transformer that we adapt from music source separation. Both behave the same way. At the size of one specialized model, the unified separator stays close to the whole models it replaces, keeping the bit error rate close on almost every pair, better on one and worse on another. On an interference absent from every training set, it does as well as the best specialized model, without anyone having to choose which one to deploy.
Be the first to rate this article.

Comments & Questions (0)

Captcha
Please be respectful — let's keep the conversation friendly.

No comments yet

Be the first to comment!