Blind Signal Classification via Spare Coding
MIT Lincoln Laboratory Lexington United States
Pagination or Media Count:
We propose a novel RF signal classification method based on sparse coding, a popular technique for image recognition in machine learning. We treat sparse coding as a configurable framework and employ a convolutional sparse coder that extracts the maximal similarity from samples of an unknown received signal against an over complete dictionary of matched filter templates. Such dictionary can be either generated or learned via unsupervised algorithms. Under this approach, we can achieve blind signal classification with no prior knowledge about signals e.g., MCS, pulse shaping in an arbitrary RF channel. Since modulated RF signals undergo pulse shaping to aid the matched filter detection by a receiver for the same radio protocol, we can exploit variability in relative similarity against the dictionary atoms as the key discriminating factor to build our classifiers. We present empirical validation of the proposed classification method. Our results indicate that we can separate different classes of digitally modulated signals from blind sampling with 70.3 recall and 24.6 false alarm at 10 dB SNR. If a labeled dataset were available for supervised classifier training, we can enhance the classification accuracy to 87.8 recall and 14.1 false alarm.
- Radio Communications