Skip to content

FFT Pre Processing for RadioML CNN #46

Description

@TanujKS

Training a CNN on just the raw I/Q values with only 128 timestamps might be the reason we are only getting ~50% accuracy. Decomposing signals into frequency bins and training on that might improve performance. Example pipeline:

I/Q -> window (Hann) -> FFT -> power -> log/normalize -> 1D CNN

Metadata

Metadata

Assignees

Labels

No labels
No labels

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions