Introduction

This article introduces the various backhaul technologies available for 5G networks, focusing on E-band wireless RF links and how they support the continued deployment of 5G networks around the world. We will perform a technical analysis of the system requirements necessary for E-band technology. We will then map the results to the physical radio design while gaining insight into the millimeter wave (mmW) signal chain.

5G Network Topology

With the successful advancement of 4G Long Term Evolution (LTE) technology, 5G networks have begun to be deployed on a large scale around the world. Figure 1 shows the topology of a 5G network to help us clearly understand the radio network from access to backhaul. The topology depicts four scenarios, each of which is connected back to the core network through a separate connection.

User equipment (UE) such as mobile phones and 5G wireless internet will access the network by connecting to base stations (gNodeBs) in the next-generation radio access network (NG-RAN). In Figure 1, we represent gNodeBs as macro cells, small cells, 5G mmW access points, and repeaters. Macro cells and small cells cover the frequency range (FR) from 410 MHz to 7.125 GHz (FR1). 5G mmW solutions cover the frequency range from 24.25 GHz to 52.6 GHz (FR2). Macro cells have a larger coverage radius, while small cells are more numerous and easier to deploy than macro cells, but have a smaller coverage radius. Small cells are used to handle traffic in densely populated areas and to more efficiently expand network capacity or coverage without adding macro cells. 5G mmW is the latest generation of technology that can meet higher network capacity requirements and support new user experiences, such as fans watching replays on mobile devices during live sports events. There are also some examples of NG-RAN equipment that can work in FR1 and FR2 bands, such as massive MIMO radios, micro cells, femto cells, and pico cells.

Figure 1. 5G network topology, including backhaul

Figure 2. Evolution of RAN

Backhaul (also known as backhaul) or mobile backhaul refers to the transport network that connects the core network (CN) and the radio access network (gNodeB in 5G). As cell site density increases, mobile and fixed wireless backhaul becomes increasingly important due to the need for high capacity links to connect to the core network. The Ericsson Microwave Outlook 2022 report shows that by 2025, urban cell sites will require 5 Gbps to 20 Gbps of backhaul capacity per site. In Figure 1, we show wireless backhaul as both microwave (μW) and E-band (mmW) radios. E-band radios can be co-located with μW radios or replace μW radios as a higher data bandwidth option. While 5G brings new business opportunities, mobile operators are under increasing pressure to quickly deliver (time to market) high capacity, low latency, reliable, scalable, cost-optimized backhaul links in urban or rural areas.

What is the difference between a back pass, a mid pass and a forward pass?

In 5G RAN, the baseband unit (BBU) functions are divided into distributed units (DU) and centralized units (CU). How operators choose to place these devices depends on the available fronthaul interfaces and link transmission technologies, and how much processing is appropriate to be done at the edge with low latency compared to a more centralized processing approach. Figure 2 shows the architectural evolution of the radio access network. Backhaul is a core part of each solution.

▪Cell   Site RAN: Traditional configuration, where the Radio Unit (RU) and BBU functions are located at the cell site. A separate backhaul link connects to the core network.

▪Centralized   RAN (low-level split): This model allows for the centralization of parts of the network to edge sites, which can provide virtualization benefits (vBBU). Processing power is pushed down to the edge sites, with only the physical layer in the cell site, thus reducing its complexity. However, fronthaul links are now required to carry large amounts of data between the RU and the centralized BBU. This is sometimes called low-level split.

▪Split   RAN (Advanced Split): RU and DU can be co-located at the cell site or separately. This model not only provides virtualization benefits (vBBU) but also improves cost-effectiveness. CN is located separately at the edge site. This is called Advanced Split:

▪   RU and DU are co-located at the cell site, while CN is located at the edge site. This means midhaul links are required to connect the remote CN (edge ​​site) to the RU + DU (cell site).

▪   RU, DU and CN are placed separately.

Both centralized and decentralized RAN models support hardware and software implementations from multiple vendors, which should bring cost benefits to network deployment. Equipment must support interoperability (RU, DU, CU), allowing solutions from different vendors to be mixed and matched to improve efficiency. This is the core spirit of the Open RAN (O-RAN) Alliance. Previously, equipment providers’ interface solutions were proprietary and could not interoperate with equipment from other vendors.

Additionally, fronthaul and midhaul links are evolving as operators deploy them in centralized and split RAN configurations. If fiber is not available and/or the cost of installing fiber is prohibitive or fiber is not a viable option for deployment in the short term, then E-band can provide an excellent solution.

It is worth noting that there is a fundamental difference between 4G and 5G: in 5G NR, the traditional EPC (Evolved Packet Core) runs on dedicated hardware, usually located near the base station or cell tower, resulting in disaggregation. This allows individual functions to run on commercial off-the-shelf (COTS) hardware. As a result, the core network of 5G is actually more decentralized as functions move to the edge. See Figure 3. Core network functions can now be co-located at the edge, making communication faster and user latency lower. It also supports network slicing, which is the creation of virtual networks for specific application needs. For example, one slice can provide high-speed broadband, while another slice can provide machine-to-machine connectivity for the Internet of Things. In addition, this edge cloud architecture supports edge computing. Therefore, the network can set up small data centers close to the edge to support video streaming of the same content, rather than struggling to backhaul data from a central location. In general, this 5G architecture is more efficient and flexible in configuring network access, hardware, functions, and backhaul.

Figure 3. 5G network slicing

What backhaul solutions are currently available?

Fiber backhaul is the highest capacity option available to mobile network operators (MNOs). It is the mainstream small cell backhaul technology used today because fiber is available in many densely populated urban/indoor areas and small cells are used in these areas to increase coverage/capacity. Fiber has a capacity of up to 1.6 Tbps (160 signals × 10 Gbps per signal). Fiber is the highest capacity option for MNOs. However, fiber deployment is costly, difficult to procure, complex and time-consuming to plan and approve. According to GMSA, the cost of deploying fiber is approximately $70,000/km. Capital expenditure and deployment time are always factors that hinder continued growth. It is important to note that μW/mmW backhaul and fiber are complementary solutions that coexist in the network. Wireless and fiber provide operators with alternative backhaul technologies. The ideal backhaul solution needs to consider many factors, including deployment time, federal/state and city permitting, obtaining right of way, data bandwidth requirements, terrain and total cost of ownership.

μW and mmW backhaul are the mainstream backhaul technologies for macro cells at present, accounting for about 50% of macro cell backhaul links.

μW licensed band technology is powerful, easy to deploy, and relatively low cost (no need to disrupt city streets or dig trenches). It covers frequencies from 6 GHz to 42 GHz, which are well suited for medium and long-range links, with a range of up to 25 km.

The use of mmW backhaul technology in the V-band (57 GHz to 66 GHz) and E-band (76 GHz/86 GHz) has been going on for many years. However, the V-band suffers from severe oxygen absorption and significant signal attenuation occurs at 60 GHz. In addition, countries have different regulations for the use of this band. Some countries license part of the spectrum for backhaul, while others leave it for unlicensed use. Europe and the United States are regions that allow unlicensed use, and regulations are being developed to reduce the probability of interference for different configurations. However, the V-band is still unreliable in providing high-quality backhaul. Its use is expected to be mainly unlicensed short-range indoor and outdoor coverage solutions (WiGig). The E-band provides a wider bandwidth solution with lower signal attenuation, which can achieve high-availability links.

So why hasn’t the E-band been used extensively in networks in the past? In 4G networks, mmW backhaul technology was not fully utilized given the available bandwidth capacity and was only used in certain scenarios, so most wireless backhaul was implemented using the licensed μW band (6 GHz to 42 GHz). With the explosive deployment and densification of 5G networks, the situation has changed, and now 10 Gbps or higher backhaul capabilities are required.

So what are the key benefits of using the E-band, and how does it compare to fiber and μW? The E-band offers two 5 GHz spectrum bands: 71 GHz to 76 GHz and 81 GHz to 86 GHz. These bands are subdivided into multiple 250 MHz channels. A major advantage of the spectrum allocation is that it can be used for either time-division duplex or frequency-division duplex links. Capacity is also not an issue, as the maximum amount of data that can be transmitted in a licensed E-band point-to-point link is greater than 60 Gbps 1 . The E-band also has the potential to be used for point-to-multipoint systems, which will further increase the available backhaul data bandwidth. The channel capacity is significantly increased compared to traditional μW radios. Due to frequency availability issues, traditional μW radio links have a capacity of only about 2.4 Gbps. In addition, E-band antennas focus the electromagnetic energy into a very narrow beam of energy (e.g., only a 1-degree divergence angle), so high-gain (45 dBi) and small form factor (30 cm antenna diameter) radios can be built, making them ideal for discreet placement on buildings or towers. Even with low RF transmit power, the E-band can typically support link lengths of up to 3 km2 . Table 1 compares several commonly used backhaul technologies.

Table 1. Backhaul technology comparison

Copper is the legacy technology using T1/E1 protocols. Copper cannot scale easily to provide the bandwidth required for 4G, let alone 5G. It is still an option for indoor small cells and public spaces, but operators are starting to move away from the technology. Satellite is not widely used compared to fiber or μW/mmW because the data rates are limited and latency is an issue due to the very high Earth orbit of geostationary satellites. Low Earth Orbit (LEO) satellites have improved latency and may play an increasing role, but the details are still uncertain. The main advantage of satellite is to connect rural areas where there are no alternatives. Wi-Fi is not a widely used backhaul technology except in very few emerging markets. These bands are unlicensed, so the increasing number of wireless access points will cause interference, and limited coverage is also an issue.

How does the wireless E-band link transmit data wirelessly?

The E-band uses traditional digital modulation coding, such as from BSPK to 1024 QAM. However, what are the factors that limit the link distance?

Inclement   Weather: Rain, fog, sleet, and snow can cause signal strength to attenuate in unpredictable ways, causing the signal level received at the receiver to drop, which in turn reduces the signal-to-noise ratio (SNR). It is important to note that the E-band radio link can use adaptive modulation when experiencing rain fade. This means that the link can switch to less complex modulation to prevent data loss. By reducing capacity during this time, the high-availability data link connection is maintained. In conditions of rainfall up to 100 mm/hour, ADI’s system-in-package (SiP) solution ensures 99.999% availability for a 1 km link.

▪Baseband   Capability: When operating at E-band frequencies, the baseband unit becomes the bottleneck for data throughput. A typical BBU supports 10 Gbps of data throughput, while the available spectrum can support over 60 Gbps of data throughput. The ADI E-band SiP will support modulation orders up to 1024 QAM.

▪   LO phase noise: Phase noise limits the modulation order. LO jitter results in a degradation of the signal-to-noise ratio (SNR) because the noise is added to the desired signal to be upconverted/downconverted. Analog Devices offers excellent broadband external phase-locked loop/voltage-controlled oscillator (PLL/VCO) sources, as well as E-band on-chip LO path frequency doublers and amplifiers.

Table 2 shows the expected bit efficiency and SNR requirements for various modulations supported by E-band technologies.

Figure 4. E-Band Radio Unit System Diagram (Blue = ADI Solution)

Table 2. Digital modulation coding and SNR supported by E-band technology

Are E-band radios more difficult to design than μW radios?

Surprisingly, E-band radios can leverage a large portion of current µW radio baseband card designs, including the modem core, processor, memory blocks, clock recovery/generation, synchronous 1588 circuits, and lower frequency analog front ends. This allows µW radio vendors to more easily transition into the E-band space. See Figure 4. The E-band front-end module, duplexer, and antenna are the new design blocks required to convert a µW radio to an E-band radio.

There is no doubt that 76 GHz/86 GHz design can seem daunting, as mmW design is more complex than lower frequency RF or even μW. As shown in Figure 4, the waveguide transition is now integrated as part of the ADI E-band SiP to minimize radio frequency (RF) losses at the antenna and convert to higher frequency signals. The ADI SiP eliminates the die, bonding, and epoxy assembly. The ADI SiP can be assembled using standard surface mount assembly equipment. The E-band SiP makes radio assembly similar to μW radio assembly.

With free space loss of 131 dB at 1 km4 and rain attenuation of 17 dB/km and 31 dB/km (for 99.99% and 99.999% availability, respectively), E-band link budgets can be challenging5 . Designers must carefully consider requirements such as gain, transmit power, noise figure, and IP3 to meet the backhaul requirements of 5G network operators.

Analog Devices has deep experience in μW and mmW backhaul technology. We developed E-band devices to address many of the above design and assembly challenges, making it easier for more designers to develop E-band products.

E-band: The next important option to meet 5G backhaul requirements

This article highlights how E-Band can provide higher bandwidth for 5G networks, thereby expanding backhaul options. It is an excellent complementary technology to fiber, providing operators with greater flexibility in planning deployments and balancing centralized and disaggregated RAN solutions.

Analog Devices has developed a surface mount, highly integrated SiP with baseband inputs or outputs and integrated waveguide outputs or inputs, eliminating much of the heavy lifting associated with E-band front-end design. Designers no longer need to worry about chip processing, but can instead take advantage of Analog Devices’ E-band packaging technology solutions. Analog Devices is committed to making technology easier to use for more RF/μW and mmW designers to drive the growth of this market. Part II will delve into the E-band link budget and technical details of the Analog Devices E-band SiP family of products.

About Analog Devices

Analog Devices, Inc. (NASDAQ: ADI) is a leading global semiconductor company dedicated to building a bridge between the physical and digital worlds to achieve breakthrough innovations in the intelligent edge. ADI provides solutions that combine analog, digital and software technologies to drive the continued development of digital factories, automobiles and digital healthcare, meet the challenges of climate change, and establish reliable connections between people and everything in the world. ADI’s revenue in fiscal year 2023 exceeded US$12 billion and it has approximately 26,000 employees worldwide. Together with 125,000 customers around the world, ADI helps innovators continue to surpass all possibilities. For more information, please visit www.analog.com/cn

About the Author

Andy Boyce is a system architect at Analog Devices, developing signal chain and system solutions. He has been designing RF and microwave products for wired, wireless, and defense systems for over 30 years. Andy holds a bachelor’s degree in electrical engineering from the University of Massachusetts Lowell and a master’s degree in finance from Bentley University.

Donal McCarthy is the director of marketing and business development for the Microwave Communications Group at Analog Devices in Cork, Ireland. He holds a BBA from University College Cork, an MBA from Boston College, and a degree in marketing from the Irish Management Institute in Dublin. Donal has held a number of positions including design engineer at MACOM, field sales engineer and marketing positions at Hittite, and marketing manager and director positions at Analog Devices.