Low-Resolution Digital Pre-Compensation for Photonic Terahertz Communications

This study explores the application of noise shaping (NS)-based low-resolution digital pre-compensation technique for enhancing photonic terahertz communication systems. Through simulations and experiments, we examined the principles of an NS quantizer with error feedback and integrated this NS approach into the digital signal processing workflow at the transmitter. This scheme involves digital preprocessing to suppress quantization noise, which improves communication signal fidelity. By optimizing NS parameters via numerical simulations, we achieved a significant increase in the signal-to-quantization noise ratio (SQNR) by over 9 dB with a 3-bit quantizer. We then experimentally validated the NS technique in a photonic terahertz communication system to address quantization noise issues inherent in low-resolution digital-to-analog converters (DACs), demonstrating improved transmission performance. For an 8 Gbaud PAM4 signal generated by a 3-bit DAC, the NS technique enhanced the signal-to-noise ratio (SNR) by more than 2 dB and reduced the bit error rate (BER) by over half an order of magnitude.