Digital radio-over-fiber (D-RoF) transmission technology is a favorable candidate for mobile fronthaul networks because it enables robust data transmission against noise, channel degradations, and nonlinear impairments in the link. Error-free transmission can also be achieved when the system is combined with forward error correction coding. However, the bandwidth efficiency of traditional D-RoF technology is relatively low because it applies uniform analog-to-digital converters with an ultra-high resolution and transmits the binary codes after digitization. In this work, the current progresses of 5G MFH have been reviewed and multiple digital transmission techniques have been discussed based on our previous works. The first method is about fast statistical estimation, which is computational efficient and particularly designed for wireless signals with Gaussian distributed amplitudes. Then, Lloyd algorithm will be introduced, which can re-allocate the quantization levels to achieve a best fit with the probability distributions of the wireless signal’s amplitudes. Thus, Lloyd algorithm can be applied to any kinds of wireless modulation format with a random amplitude distribution, but a trade-off needs to be considered between computational complexity and compression ratio. Differential coding-based compression techniques are also discussed, where we have proposed an adaptive low-complexity differential encoder based on least-mean-square (LMS) algorithm to further improve the compression ratio and be adaptive to the signal’s dynamic change. By jointly using Lloyd algorithm based quantizer and an LMS differential encoder, significant improvements on compression efficiency can be achieved. The Lloyd algorithm and differential coding have been jointly applied in a data-compressed mobile fronthaul testbed with a net data rate of 100 Gbit/s, which is able to comprise 45×120- MHz 5G NR carriers with lower-than 1% EVM.
A new function split option 9 based on delta-sigma modulation is proposed for the next generation fronthaul interface (NGFI). All existing low layer split (LLS) options, such as 6 (MAC-PHY), 7 (high-low PHY), and 8 (CPRI) require a complete RF layer implemented in the analog domain at remote cell sites; whereas the proposed option 9 can implement RF layer in the digital domain and moves the split point between the distributed unit (DU) and remote radio unit (RRU) into the RF layer. With the help of delta-sigma modulation, high-RF layer functions are centralized in the DU and lowRF layer distributed in RRUs. Although it splits at a lower level than conventional option 8 (CPRI), option 9 offers improved spectral efficiency and reduced fronthaul traffic than CPRI. Moreover, it implements all RF functions in the digital domain and eliminates the need of analog devices required by other LLS options, such as DAC, local oscillator, mixer, power amplifier. Therefore, option 9 enables all-digital RF transmitter and RF layer virtualization, which makes low-cost, energy-efficient, and small-footprint cell sites possible for the wide deployment of 5G small cells.