In a previous blog post we assessed the evolution of 5G wireless standards and industry expectations for the incremental deployment of 5G infrastructure in the years ahead. Continued forward progress on these fronts brings us ever closer toward a future of breakthrough bandwidth capacity and data delivery speeds. There will of course be daunting technology challenges on the pathway to 5G, but these challenges present tremendous opportunities for wireless operators to evolve our wireless infrastructure in a scalable and sustainable manner that enables continuous improvements in bandwidth, power, management and cost efficiencies.
Wireless fronthaul is a prime example of this dynamic, inviting a fresh look at the underlying network architecture extending from remote radio units (RRUs) to baseband units (BBUs). Taking a comprehensive view of wireless fronthaul trends and standards, we see a clear opportunity to combine parallel advancements in RF and optical technologies to realize a more elegant and cost effective fronthaul architecture that’s optimized to accommodate the huge increase in 5G data throughput.
CENTRALIZED NETWORK ARCHITECTURE
At the RF antenna layer, advanced phased array technologies are enabling massive MIMO configurations that multiply the capacity of antenna links leveraging anywhere from 16 to hundreds of densely clustered antennas. Meanwhile, each individual antenna is targeted to serve more and more bandwidth, from 60 MHz today to as high as 200 MHz for the 3.5 GHz frequency band, and far higher – as high as 800 MHz – for mmWave frequencies.
The data throughput enabled by this advanced RF architecture is immense, but introduces significant dilemmas about how best to partition the attendant signal processing burden between the RRU and BBU to maximize overall data throughput. If the baseband processing is integrated into the RRU via an “All-in-One BTS” approach, then RRU power consumption will rise dramatically, as will maintenance and upgrade costs. This approach runs counter to the inherent benefits of the Centralized/Cloud Radio Access Network (C-RAN) architecture, whereby the RRU design is greatly simplified and the processing intelligence instead resides at a centrally managed, maintained, and cooled BBU servicing a multitude of RRUs over long distances.
C-RAN leverages the natural efficiencies of virtualized, general purpose server infrastructure, with highly flexible, software defined programmability. With C-RAN, it’s also much easier to administer and implement algorithms to mitigate inter-cell interference and improve spectral efficiency.
CPRI vs eCPRI
Though C-RAN promises many benefits for 5G wireless fronthaul, it’s not without its challenges. Chief among them, the long distance functional split between BBU processing and RU requires an ultra high speed optical interconnect to minimize latency. The Common Public Radio Interface (CPRI) protocol, established in 2003, is commonly employed for this purpose today over distances of tens of kilometers, but has been bandwidth constrained in the absence of low cost, high speed optical connectivity solutions.
Published in 2017, the CPRI Over Ethernet (eCPRI) specification aims to address this issue in part by moving a portion of the processing workload from the BBU over to the RU, where the dataflow can be pre-processed in a manner tailored for the requirements of specific 5G uses cases including fixed wireless, massive broadband, and ultra low latency applications. But here again, this approach runs counter to the core value proposition of C-RAN and the centralization of processing workloads and network administration at the BBU.
5G network split options under consideration (image courtesy of eCPRI Overview/IEEE)
As an alternative approach to eCPRI, data compression techniques are now being developed for implementation at the RRU, targeted to accommodate the extreme data throughput introduced with massive MIMO while ultimately reducing fronthaul bandwidth requirements to a level commensurate with today’s LTE MIMO architectures – a CPRI data rate that’s serviceable with 100G optical interconnects. By compressing the dataflow from the RRU to the BBU, the combined benefits of CPRI and C-RAN can be readily achieved, enabling BBU-level consolidation of baseband processing and management operations.
MACOM’s innovative 100G single-wavelength PAM4 DSP and advanced silicon photonics technology deliver a high-performance, cost-effective 100G optical solution for C-RAN networks. Where previously the CPRI protocol enabled streamlined C-RAN architectures but fell short in data delivery speeds, the mainstreaming of 100G optical connectivity has bridged this gap, thanks also to the volume scale deployment of 100G connectivity in cloud datacenters, and the resulting reduction in 100G cost structures. Wireless operators are now well positioned to exploit the speed and cost efficiencies of 100G interconnects for wireless fronthaul in preparation for the 5G data deluge, while enabling a flexible C-RAN architecture designed for seamless scalability into the future.
MACOM’s leadership across the RF and optical networking domains gives us unique visibility into the technologies, trends and standards driving the development and deployment of 5G connectivity. To learn more about MACOM’s 5G solution portfolio, visit https://www.macom.com/applications/wireless-networking-and-communic/5g
All financial guidance projections referenced in this post were made as of the publication date or another historical date noted herein, and any references to such projections herein are not intended to reaffirm them as of any later date. MACOM undertakes no obligation to update any forward-looking statement or projection at any future date. This post may include information and projections derived from third-party sources concerning addressable market size and growth rates and similar general economic or industry data. MACOM has not independently verified any information and projections from third party sources incorporated herein. This post may also contain market statistics and industry data that are subject to uncertainty and are not necessarily reflective of market conditions. Although MACOM believes that these statistics and data are reasonable, they have been derived from third party sources and have not been independently verified by MACOM.