Boosting 5G Through Ethernet: How Evolved Fronthaul Can Take Next-Generation Mobile to the Next Level

Abstract

The version in the Kent Academic Repository may differ from the final published version. Users are advised to check http://kar.kent.ac.uk for the status of the paper. Users should always cite the published version of record.

A Survey of the Functional Splits Proposed for 5G Mobile Crosshaul Networks

“…It is important to note, that IEEE 1914.1 complies with 3GPPdefined partitioning schemes, but does not aim at defining them. [118], [119] defines the scope of the P1914.1 project to specify both an Ethernet-based architecture for the transport of mobile fronthaul traffic, including user data traffic and management and CP traffic, and requirements and definitions for fronthaul networks, including data rates, timing and synchronization, and QoS. IEEE 1914.1 defines a two-level fronthaul architecture that separates the traditional RRU to BBU connectivity in the C-RAN architecture into two tiers, via interfaces called NGFI-I and NGFI-II.…”

“…This creates a possibility to use the already established Ethernet network. The standard defines packetization of IQ samples in both the time domain and frequency domain, using mappers to transfer existing radio transport protocols over Ethernet, transferring native IQ data as well as allows for externally-defined mappers [122], [119]. The IEEE 802.1 CM draft standard looks into Time-Sensitive Networking for Fronthaul, by defining profiles to select features, options, configurations, defaults, protocols and procedures to build time sensitive networks [123].…”

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

“…Digital transport techniques in the fronthaul section of the Radio Access Network (RAN), traditionally based on the transportation of digitized radio samples [1], cannot scale to the signal bandwidths employed by the 5 th generation (5G) mobile network [2], [3]. Furthermore, the use of multiple antenna techniques such as massive Multiple-Input Multiple-Output (mMIMO) means that bit rates over the fronthaul not only scale with the number of radio access technologies, signal bandwidths, carriers and sectors but also with the number of (spatial) radio streams, making the use of such transport techniques exceptionally challenging [2][3][4][5][6][7][8].…”

“…The remaining RAN protocol stack processing is split between the DU and a Radio Unit (RU). 3GPP has not reached a consensus on the split point in RAN functions between the DU and RU, but possible options within the baseband Physical (PHY) layer have been proposed [4][5][6][7][8][9][10]. The fronthaul is the transport network for this F2 interface between the functionally decomposed DU and RU.…”

Testbed Verification of New Fronthaul Technology for 5G Systems

“…Apart, from the general loss of centralization advantages for NFV, distributed beamforming such as joint transmission/reception with coordinated multipoint (CoMP), centralized coordination of interference cancellation, all become difficult or impossible to achieve as the functions required for them are distributed to remote units [6]. Now, it is generally accepted that there are two types of new functional split for the Radio Access Network (RAN) [7]. A higher-layer split, which is outside of the HARQ timing loop, with relatively relaxed latency requirements on the transport over this fronthaul segment.…”