KNect365 is part of the Knowledge and Networking Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 3099067.


The Role of Microwave and Millimeter Wave Technologies in 5G Mobile Networks

Microwave technologies will have an important role to play in 5G networks explains Huawei's Leonida Macciotta.

5G is the highly anticipated technology that is expected to change the way we live by connecting everything around us.  According to the ITU, 5G falls into three main categories. 1) Enhanced  Mobile Broadband (eMBB), 2) Ultra-Reliable and Low Latency Communications (uRLLC) for critical machine communications; 3) Massive Machine –to Machine (M2M).  Each application requires a slightly different set of parameters.  To facilitate the “enhanced broadband everywhere” concept that comes with 5G, gigabit (i.e. 10 Gbps) levels of capacity will be required, which has some wondering if microwave has the wherewithal to backhaul 5G cell sites.

As 5G deploys, the number of mobile sites will multiply along with capacity, complexity, and stringent latency requirements.  While it is true, that fiber will be used if it is available, it is also true, that fiber won’t be able to reach every site, and other technologies will need to be considered.  This is where microwave will be essential.  In order to support 5G, the main key areas are discussed below.


While the industry agrees that more capacity will be required for a 5G mobile network, exactly how much capacity will be needed at each point within the network.  Capacity requirements vary greatly depending on the location of the cell site.  Cell sites at the network’s edge, where the majority of cell sites are located, will require the least amount of capacity.  Capacity will increase as traffic moves toward the core of the network and aggregated onto fewer sites.  Thus, while multi gigabit capacities are expected, they are expected at only a few very high end sites.  The common understanding currently seems to be that 1 Gbps of capacity or less will be needed at Macro Sites through 2020, with the most extreme sites calling for 10 - 20 Gbps in 2025.  20 Gbps capacity is already a reality with equipment operating in the 80 GHz band (band E) and aggregation techniques, and capacities up to 100 Gbps are expected to be reached with equipment operating in the new 150 GHz band (band D).

A traditional solution to increase capacity over a radio link is bonding multiple channels. These channels may be in the same frequency band, providing a linear increase in capacity, or exploit the different characteristics of different frequency bands, namely the amount of available spectrum (maximum in E and D band) and the ability to withstand attenuation by rain (best in lower frequencies, from 6 GHz to 42 GHz, usually referred to as “traditional microwave bands”).

Combining transmission over a lower band (high availability, relatively lower capacity) with transmission e.g. in E band, allows to exceed 10 Gbps over distances of about 10 km. Considering that two thirds of microwave links is less than 10 km long, such techniques allow to cover most of the required use cases.  Carrier aggregation not only saves operators money by enabling them to leverage existing links when upgrading backhaul capacity in their current IP microwave network, but it also helps operators build their 5G backhaul by economically transporting multiple gigabit capacity from challenging sites that are not accessible by fiber.


Another key requirement for 5G mobile backhaul network is low latency, especially in uRLLC applications like automated driving, collaborative robots, and remote surgery.  4G legacy networks are not able to support these critical applications, thus opening up an additional revenue stream for operators willing to make the 5G network investment.

The latency requirements go from 10 ms (round trip) for the eMBB services, to less than 1ms for the uRLL services.

To meet those requirements, all network levels and segments are impacted, from the packet routing/forwarding to the physical level of the transmission. At packet processing level, enabling L3 routing capability all the way to the tail site, the forwarding path becomes shorter and thus lowers latency to around 1-2 milliseconds. Additional special provisions at the L2 switch and Ethernet interface level allow to keep a predictable and stable delay performance.

At microwave link level, a latency as low as 50 microseconds can be achieved over a single hop using traditional microwave bands (6 - 42 GHz).  This means that even with several hops, end to end latency still comes well within the acceptable range for 5G backhaul specifications.


Cell densification not only lowers latency but also increases capacity.  Thus, densification is achieved by overlaying macro cells with small cells.  Future small cell topologies will use a combination of unlicensed (i.e. sub 6 GHz or 60 GHz) radios, 80 GHz and 150GHz radios, traditional microwave radios, and/or fiber. One of the advantages to use very high frequency radios (like D and E band) is that the antenna radiation pattern is very narrow, allowing to pack a much higher number of links per area compared to lower frequencies (it has been calculated that up to 5x link density increase is reachable in practical situations).

To facilitate the street-level network topology, microwave radios are being designed in all outdoor, small compact form factors that easily blend with the urban landscape.  Although outdoor small cells have not been widely deployed, operators are already taking advantage of this new, smaller form factor.

Street level cells will be built using very short, high capacity hops using unlicensed 60 GHz spectrum.  In some regions where 70-80 GHz is inexpensive, then 70-80 GHz can be used as well in hotspot sites.  In an effort to reduce clutter on the street, radios will be hung on street lights, traffic poles, and bus stop signs.  To eliminate any problems with sway or NLOS that may be inherent with alternative masts, a wide beam antenna will be used and radios will be placed strategically to bounce signals off each other to work their way around buildings and other obstructions.

Additionally, a mesh network topology may be used, which will create multiple paths a signal can take if a hop is blocked by a passing bus or a blooming tree.  The street level signal will be aggregated and carried to the top of a building using a 70-80 GHz hop.  Once again, in lieu of towers, operators will take advantage of already existing infrastructure - hanging radios on the sides of buildings and carefully disguising them so they are aesthetically pleasing to the community.  Once the aggregated signals hit the rooftop, the signal will be carried to another cell site via microwave link and absorbed into the greater macro network.  Alternatively, the radio on the top of the building may be connected via fiber to the building, and will be absorbed into the network at that point.  Either way, a multitude of technologies will be required, deployed, and managed in a complex architecture, further highlighting the need for SDN enabled microwave.

Spectrum and Spectrum Efficiency

In order to achieve the increase in transport capacity required by 5G, it’s very important to open new frequency bands for backhaul, and to use the existing ones more efficiently.

The new bands are essentially the E band (10 GHz free spectrum around 80 GHz) and the D band (around 30 GHz free spectrum around 150 GHz). The huge amount of available spectrum makes it possible to achieve 10 Gbps on one channel today, and in perspective 100 Gbps in a few years, just in time for when 5G will require such capacities.

In terms of spectrum efficiency, the available tools include using higher and higher modulations (12 bits/symbol today, up to 16 bits/symbol in the future), MIMO (quadrupling the spectral efficiency) and interference cancellation systems (allowing a much higher reuse of frequency and denser networks).

Flexible OAM by SDN

As more options are added to the network, the network becomes more complex, requiring a high degree of flexibility in a dynamic network often built from multiple manufacturers.  Software defined networks (SDN) help make the complex 5G network easier to manage by implementing a variety of software management protocols that automatically manage the network’s traffic based on dynamic physical attributes of the network.  Although SDN for microwave is a relatively new concept, forward thinking carriers moving toward 5G are doing so with SDN.   Vodafone has stated that “5G has key components that really can’t happen without SDN and NFV and those include low latency applications, self-healing networks and an ability to manage mobile applications at the very edge of the network using a technology called Network Slicing, used to create multiple dedicated end-to-end virtual networks.”  Although Vodafone is speaking about the greater network, SDN saves the carrier money, and SDN in the microwave is no different.

SDN on a microwave link can automatically re-route traffic in peak times with congested links in order to keep data traffic flowing.  It also can make the network more efficient and keep operational costs in check by managing link usage and adjusting traffic to one or more links.  It has been estimated that by moving traffic off a spare link and adjusting modulation and transmit power during off peak times, operators can realize a 15-20 percent savings in power consumption costs, and over an eight year period, it can save the operator 50 percent of their capital investment.


5G promises an exciting future, and with the ongoing advances in microwave, it will be a key component of the network.  Through the use of new spectrum, SDN, and lower latency achievements, microwave vendors are demonstrating that the technology is able to meet the 5G backhaul specifications economically and efficiently for the foreseeable future.

Get articles like this by email