End-to-End Radio Traffic Sequence Recognition with Deep Recurrent Neural Networks
This addresses the problem of automated radio traffic analysis for communication systems, offering a novel end-to-end approach that bypasses traditional demodulation methods.
The paper tackled the problem of recognizing application layer traffic types from raw radio signals without expert demodulation, achieving the ability to both classify and generate complex protocol sequences using deep recurrent neural networks.
We investigate sequence machine learning techniques on raw radio signal time-series data. By applying deep recurrent neural networks we learn to discriminate between several application layer traffic types on top of a constant envelope modulation without using an expert demodulation algorithm. We show that complex protocol sequences can be learned and used for both classification and generation tasks using this approach.