Long-range photonics-aided 17.6 Gbit/s D-band PS-64QAM transmission using gate recurrent unit algorithm with a...

Area of Science:

• Optical Communications
• Wireless Communications
• Machine Learning

Background:

• Long-range D-band wireless transmission faces limitations due to absorption loss and system nonlinearities.
• Exploring m-QAM formats is crucial for improving spectrum efficiency and SNR in D-band systems.
• Nonlinearity in photonics-aided millimeter-wave systems necessitates advanced compensation techniques.

Purpose of the Study:

• To investigate the effectiveness of machine learning algorithms for nonlinear compensation in D-band wireless transmission.
• To propose and evaluate a novel Gate Recurrent Unit (GRU) algorithm with complex QAM input for improved receiver sensitivity.
• To compare the performance of complex-valued neural network (CVNN), single-lane Long Short-Term Memory (SL-LSTM), and single-lane Gate Recurrent Unit (SL-GRU) for D-band signal recovery.

Main Methods:

• Implementation of adaptive deep learning methods, including CVNN, SL-LSTM, and SL-GRU, with complex QAM input.
• Experimental setup for 135 GHz wireless transmission over 4.6 km.
• Evaluation of signal recovery precision and transmission capacity for different modulation formats (QPSK and PS-64QAM).

Main Results:

• Successful wireless transmission of 135 GHz 4Gbaud QPSK and PS-64QAM signals over 4.6 km.
• Demonstration of improved receiver sensitivity using the proposed GRU algorithm.
• CVNN equalizer is optimal for QPSK recovery, while SL-GRU is best for PS-64QAM in long-range D-band transmission.
• Achieved effective data rates up to 17.6 Gbit/s.

Conclusions:

• The combination of high-order modulation and NN-supervised algorithms with complex input shows significant promise for future 6G mobile communications.
• SL-GRU offers superior performance for PS-64QAM recovery in long-distance D-band wireless links.
• Deep learning-based nonlinear compensation is critical for enabling high-capacity, long-range D-band communication.