This is the sixth article in an eight-part series exploring 5G. The series will explain key terms and technologies and provide an overview of current and future applications for 5G connectivity.
The articles were originally published in an e-magazine, and have been substantially edited by Wevolver to update them and make them available on the Wevolver platform. This series is sponsored by Mouser, an online distributor of electronic components. Through their sponsorship, Mouser Electronics supports a more connected future fuelled by knowledge and innovation.
The fifth-generation (5G) of what used to be called cellular radio is like none before it. It encompasses not just smartphones and tablets but almost everything that can benefit from being connected to other things and the Internet. For the first time, connecting businesses is as important as connecting consumers, and while mobility remains the core focus of the 5G operating environment, 5G will also serve many fixed applications. Whether catering to mobile or fixed applications, the 5G infrastructure will still need base stations—orders of magnitude more of them than currently exist—but deploying these base stations may be the most challenging task in developing a complete 5G ecosystem.
The impact of 5G on the current wireless infrastructure affects all three major use cases: Traditional mobile, called enhanced mobile broadband (eMBB); massive machine-type communications (mMTC) Internet of Things (IoT) environments; and the newest and most demanding use case, ultra-reliable low-latency communications (URLLC). Serving all three of these use cases requires a change from a hardware-based network architecture to one that’s virtual and based on network function virtualization (NFV), software-defined networking (SDN), processing and analytics at the network’s edge, network slicing, and other technologies new to the wireless infrastructure.
The transition from hardware- to software-based (i.e., virtual) networks affects base stations of all sizes. The two “virtualizers” of this new paradigm—NFV and SDN—perform similar functions because they provide far greater network control using less hardware, thus reducing or eliminating changes to hardware as the network evolves. The capacity of these virtualizers also means less hardware may be necessary.
The benefits of network virtualization are now being realized as virtualization empowers software to perform functions in the digital domain rather than through analog and digital hardware, as has been the case before. Still, this shift to virtualization will quite obviously require substantial computational resources, a point that has not been overlooked by Arm Limited (Arm®), an organization best known for its architectural designs for smartphones, network processors, and embedded processors.
Arm’s first computing ecosystems tailored exclusively for servers and infrastructure design are the Neoverse™ series of processors. Each processor in the series includes not just a central processing unit (CPU) core but an interconnected scheme, making it possible to scale up to many cores. Neoverse processors will likely find a home at the network’s edge, where high-performance computing is increasingly becoming necessary. These processors are also well suited for 5G base stations, which will ultimately handle much more data than 4G Long-Term Evolution (LTE) networks but with greater DC power constraints.
An entirely different approach will be required for the technology dedicated to generating and receiving signals over the air. For example, techniques such as massive multiple-input, multiple-output (MIMO) that rely on digital signal processing (DSP) still need to convert the analog signal to a digital form. Fortunately, major advances are being made to address this need from lower frequencies to the microwave, and millimeter-wave regions, with semiconductors that generate and receive RF signals to and from the antennas.
RF Output Power
For the past two decades, base station RF power amplifiers (PAs) have used lateral double-diffused metal-oxide semiconductor (LDMOS) field-effect transistors to generate RF power in macro-base stations. This technology has advanced dramatically over the years with respect to the amount of power a single device can generate (currently, about 2 kilowatts [kW]) and use efficiently, beating every other potential competitor. But, LDMOS has a maximum useful operating frequency of about 4 gigahertz (GHz), which eliminates its use at most of the new frequencies proposed for 5G networks. The technology that will take the mantle is gallium nitride (GaN), which, in the less than 15 years it’s been commercially available, has cemented its place as the next big thing in RF power.
For example, a GaN semiconductor die can produce 10 times the RF power output per unit of die than gallium arsenide (GaAs), with higher efficiency, higher-voltage operation, and superior thermal characteristics. GaN on silicon carbide (SiC) substrates offers better performance than GaN on silicon substrates, but it also costs more. This means that the latter will find a home in cost-critical small cells, where it will compete with GaAs, the predominant power semiconductor technology that has been used in smartphones for many years.
At millimeter-wave frequencies, where high RF output is difficult to achieve in a solid-state device, silicon germanium (SiGe), complementary metal-oxide semiconductor (CMOS), and GaAs will be the key technologies. All these technologies, except LDMOS, will also be important for either RF power generation or receiving applications, such as low-noise amplifiers (LNAs).
RF Front Ends: Packaging Functional Integration
The 5G infrastructure will require that manufacturers of microwave and millimeter-wave devices increase their functional integration to reduce costs, complexity, and size and meet the needs of small cell base stations (as well as end-user devices). Packaging technology will be an essential ingredient in achieving this goal because although 4G increased the number of bands that a radio must accommodate to about 30, 5G will increase this number to 40 and perhaps more. Accommodating all these bands in a single device is unlikely, especially considering each band has different characteristics and thus requires different technologies. Nonetheless, each band segment, from sub-1GHz to millimeter wavelengths, must be as highly integrated as possible.
To create these highly integrated RF devices, manufacturers of RF front ends (RFFEs) are taking a variety of approaches. Let’s see what approaches Analog Devices, Skyworks Solutions, and Qorvo are taking to fabricate products that accomplish this feat.
The good news for manufacturers and carriers is that all the necessary technologies to fully implement the 5G infrastructure won’t be needed all at once, so developments can continue apace as networks are deployed in the number that is essential to blanket coverage areas. Considering this flexibility in the 5G infrastructure, all signs point to 5G rolling out close to predicted timelines.
Verizon™, AT&T®, and T-Mobile® continue to add more and more cities to the list of where 5G wireless is available in the United States and in Europe, as of June 2020 commercial 5G has been deployed in 14 countries through service providers including Three, T-Mobile and Vodafone.
This article was originally written by Barry Manz for Mouser and substantially edited by the Wevolver team. It's the sixth article of an eight-part series exploring 5G. Future articles will explore how 5G differs from existing technology and how the potential of hyper-connectivity will be applied in entertainment, smart cities, and industry and how to get there.
Article One gives an overview of 5G.
Article Two introduces key terms and technologies.
Article Three discusses how systems engineers can evaluate the viability of 5G in the existing connectivity ecosystem.
Article Four examines the relevant standards associated with 5G.
Article Five showcases the radical applications 5G will enable.
Mouser Electronics is a worldwide leading authorized distributor of semiconductors and electronic components for over 800 industry-leading manufacturers. They specialize in the rapid introduction of new products and technologies for design engineers and buyers. Their extensive product offering includes semiconductors, interconnects, passives, and electromechanical components.