PARGAR Graph Attention Receiver Improves Sparse-Pilot OFDM
Original title: PARGAR: Pilot-Aware Graph Attention for Sparse-Pilot Neural OFDM Receivers
PARGAR yields lower BER than DeepRx in selected sparse-pilot 5G scenarios.
PARGAR is a graph-attention neural receiver that represents the OFDM grid as a graph with pilot-anchor links. It uses a channel-scaled pilot-anchor rule based on coherence time and bandwidth. Under Sionna simulations, it yields lower BER than DeepRx in sparse DMRS layouts while remaining competitive with dense pilots.
- Graph-attention receiver for OFDM resource grid
- Lower BER than DeepRx in sparse-pilot scenarios
- Competitive under dense pilots
- Injects pilot-aware inductive bias
From the original article
1 min read at arxiv.orgReliable channel estimation and symbol detection in 5G New Radio (NR), as well as emerging 6G radio systems, rely on demodulation reference signals (DMRS). The density and placement of these pilots on the orthogonal frequency-division multiplexing (OFDM) time--frequency resource grid affect both receiver reliability and spectral efficiency.
By Sajedeh Norouzi · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.