DeepRx
A CNN-based neural receiver proposed by Nokia Bell Labs that maps a resource grid containing DMRS end-to-end to LLRs of coded bits, relying on local convolutional receptive fields to extract time-frequency features.
Timeline (last 90 days)
CNN-based neural receivers (such as DeepRx) rely on local receptive fields and struggle to fully exploit distant pilot information under sparse pilots.
The paper evaluates using OFDM simulations generated by Sionna, with the CNN-based DeepRx receiver as the comparison target, covering three DMRS layouts—dense, two-symbol sparse, and one-symbol sparse—and also examines channel mismatch scenarios.
In the SISO uncoded setting, PARGAR with the channel scaling rule achieves a lower bit error rate (BER) than DeepRx in some sparse pilot scenarios while remaining competitive under dense pilots.
DeepRx is a CNN-based neural receiver proposed by Nokia Bell Labs that maps the received resource grid (including DMRS) directly to log-likelihood ratios (LLRs) of coded bits and is trained end-to-end.
All Facts (4)
CNN-based neural receivers (such as DeepRx) rely on local receptive fields and struggle to fully exploit distant pilot information under sparse pilots.
50%UnverifiedThe paper evaluates using OFDM simulations generated by Sionna, with the CNN-based DeepRx receiver as the comparison target, covering three DMRS layouts—dense, two-symbol sparse, and one-symbol sparse—and also examines channel mismatch scenarios.
50%UnverifiedIn the SISO uncoded setting, PARGAR with the channel scaling rule achieves a lower bit error rate (BER) than DeepRx in some sparse pilot scenarios while remaining competitive under dense pilots.
50%UnverifiedDeepRx is a CNN-based neural receiver proposed by Nokia Bell Labs that maps the received resource grid (including DMRS) directly to log-likelihood ratios (LLRs) of coded bits and is trained end-to-end.
50%