Ultra-Low-Latency, High-Fidelity Analog-to-Digital-Compression Radio-Over-Fiber (ADX-RoF) for MIMO Fronthaul in 5G and Beyond

Abstract

Wired-wireless convergence technology such as radio-over-fiber (RoF) is regarded as a promising candidate to deliver broadband wireless data from/to edge cloud to/from antenna sites, a notable example being the fronthaul (FH) in centralized/cloud radio access network (C-RAN). In 5G and beyond, the latency, bandwidth and fidelity requirements on FH pose great challenges to the RoF schemes. In this article, we focus on analog-to-digital-compression RoF (ADX-RoF) scheme based on low-latency MIMO data compression, which has the potential to reduce FH data rate by ~90% compared with traditional digital RoF (CPRI) while still maintaining signal fidelity for high-order radio modulation. To verify the practicality of ADX-RoF in the latency-sensitive FH scenario, real-time hardware demonstration with latency evaluation is indispensable. We propose and design a real-time ADX, which achieves low-latency and high-throughput together with high-fidelity. Enabled by the ADX prototyped on a single-chip field-programmable radio platform (i.e., Xilinx RFSoC), we experimentally demonstrate 16-channel MIMO radio reception, real-time ADX processing and RoF transport. With <; 500 ns one-way ADX latency, less than 1.5% average EVM is achieved for 1024QAM, 61.44 MHz NR-class signals at a compression ratio of 13.3%. Negligible fiber-induced performance penalty is observed after 40 km transport. The results highlight the attractiveness of ADX-RoF for future FH.