To find some answers, Heavy Reading launched its first-ever 5G Transport Market Leadership Study with support from Ericsson, Fujitsu, Juniper and VIAVI Solutions. The analysis is anchored by a global survey of 104 qualified network operator respondents. Based on the new research, this blog summarizes some of the key issues and priorities in 5G fronthaul, midhaul and backhaul networks, collectively known as "Xhaul." This is the second of two blogs in the series on 5G transport.
The centralized RAN (C-RAN) creates a new transport network segment -- the fronthaul network -- with new performance requirements and protocols. The CPRI Consortium created eCPRI specifically to economically meet the higher bandwidth requirements of 5G fronthaul compared to 4G fronthaul. Given the importance of C-RAN for 5G and the role that eCPRI will play in fronthaul transport, it's not surprising that eCPRI topped the list of most important technologies for 5G in the Heavy Reading survey. Forty-eight percent of respondents picked eCPRI as "highly important."
Also scoring at the top of the technology list was Optical Transport Network (OTN). This top ranking was not limited to a specific geographic region; rather, it spanned all regions (including in North America). Here, OTN is largely a proxy for wavelength-division multiplexing (WDM) wavelengths (which use OTN framing). 5G will be the main driver in telecom for WDM expansion into access networks (or Xhaul). What is not clear from this question is the extent to which operators will want switched OTN in their access and aggregation networks. Heavy Reading gets mixed opinions from one-on-one discussions with operators.
Finally, we note a respectable showing from 10Gbit/s microwave as a 5G transport technology. The microwave option was selected as "highly important" by 35% of respondents and ahead of some other strong technology options. While fiber is the clear first choice for transport, it will not always be available or economical. Heavy Reading's market research continues to show that high data rate microwave will play an important role in 5G transport.
Toward RAN interoperability
Historically, the RAN was a closed network in which radio unit (RU) and baseband unit (BBU) products and functions were supplied by the same vendor. With 5G, however, the RAN is distinctly moving toward more open interoperability, and the trend is being driven directly by operator demand. In the Heavy Reading survey, 77% of respondents reported that RAN interoperability between RUs and BBU equipment was at least "very important" to them, with 22% of respondents reporting that RAN interoperability was a "critical" requirement that must be supported by all vendors. These are important findings that underscore the growing importance of operator-led interoperability initiatives, such as the O-RAN Alliance. While many vendors are coming around to RAN interoperability, some are still hoping it's a passing fad. The research suggests it is not.
5G RAN decomposition and centralization lead to another new Xhaul transport segment -- the midhaul segment that connects distributed unit (DU) processing to the central unit (CU). While distinct, midhaul also shares much in common with the backhaul network in terms of requirements. Operator survey data shows that several factors will drive midhaul/backhaul network upgrades, but the need for greater capacity is -- by far -- the chief reason. In Heavy Reading's survey, 69% of respondents selected capacity as the primary reason -- far ahead of the second option, latency.
5G New Radio (NR) ultimately promises a 10x (or greater) increase in end-user capacity. This increase is not confined to the radio network alone, as it reverberates throughout the Xhaul network. Current 4G backhaul networks are based on 1Gbit/s Ethernet connectivity. While 5G may not fill a 10Gbit/s midhaul or backhaul link on Day 1 of deployment, it will require greater than 1 Gbit/s per port. As per-port rates increase from 1 Gbit/s to 10 Gbit/s (or higher), port density must also increase.
Second only to greater capacity, low latency was a primary reason for midhaul/backhaul network upgrades (selected by 40% of respondents). Fiber connectivity, short links and edge compute are not effective if packets get delayed in processing or in buffers. Therefore, the packet layer must also be upgraded for the most efficient processing.
Readers interested in further data points and analysis from this survey can access the full survey report, Operator Strategies for 5G Transport. Additional information can also be found in the related archived webinar, Meeting the Xhaul Challenge, in which Heavy Reading and study sponsors expand on the 5G transport topic.
— Sterling Perrin, Principal Analyst, Heavy Reading
This content is sponsored by Ericsson, Fujitsu, Juniper and VIAVI Solutions.