Comparison of digital signal processing approaches for subcarrier multiplexed 5G and beyond analog fronthaul

Analog fronthaul transport architectures with digital signal processing at the end stations are promising, as they have the potential to achieve high spectral efficiencies, increased flexibility, and reduced latency. In this paper, two digital techniques for frequency-domain multiplexing/demultiplexing large numbers of channels are contrasted: one operates on the pre-inverse fast Fourier transform (IFFT) “frequency-domain” samples, while the other does so on the post-IFFT “time-domain” samples. Performance criteria, including computational complexity and sampling rate requirements, are used in the comparison. Following modeling and simulation of the techniques, implemented within a radio-over-fiber transport architecture, error vector magnitude performance estimates are obtained. These results show that each technique has performance advantages under specific channel transport scenarios.