Sub THz OFDM receiver improves error rates with channel and noise estimation

PACT: A Decoder-Confidence-Guided Two-Round Receiver for Joint Channel and Phase-Noise Estimation in Sub-THz OFDM

Machine Learning

Summary

Signals in sub-terahertz communication systems are affected by changes in the channel and noise from oscillators, making it hard to decode data accurately. The authors propose a new receiver design that first estimates the channel and noise conditions, then uses this information twice to refine the decoding process. This approach significantly lowers errors compared to previous learning-based methods. Their tests show it works well even when configurations or environments change.

What this means in practice

Authors

Ruilin Wang, Xiaodai Dong

Abstract

Frequency-selective fading and oscillator phase noise (PN) jointly impair coded sub-THz OFDM reception. We propose a decoder-confidence-guided two-round receiver combining a {pilot-aware complex transformer} (PACT) estimator, banded linear minimum mean-square error equalization, and neural normalized min-sum decoding. PACT jointly estimates the frequency-domain channel and effective time-domain PN through interacting complex-valued branches. Decoder-derived pseudo-pilots and continuous constellation-aware confidence guide second-round refinement through PACT inputs and attention, with parameters shared across rounds. At $E_b/N_0=30$~dB, the PACT receiver achieves approximately $70\%$ and $57\%$ lower code-block error rates than the best-performing learning-based baseline in the two benchmark settings, respectively. Frozen-weight tests further demonstrate applicability across channel/PN models, carrier frequencies, and FFT/bandwidth configurations.