The outbreak of the new crown pneumonia has brought challenges to the global supply chain. But before that, the RF and microwave semiconductor industries have already faced tremendous resistance. The cellular communications market, especially handheld devices, accounts for more than 50% of compound semiconductor revenue. For more than ten years, this application has been a strong driving force in the industry, but now some follow-ups are weak. The revenue of radio frequency GaAs equipment declined in 2019, mainly due to the decline in smartphone shipments. Nevertheless, the future of the compound semiconductor industry is still bright. This optimistic estimate mainly stems from 5G networks and equipment. This new standard is expected to become the growth engine of the entire semiconductor industry.
5G market
Since 2019, wireless operators have been deploying 5G networks and equipment, so people should be familiar with the three cores of the 5G vision. Figure 1 simply shows its main components and the functions that these three projects can achieve. The challenge that operators and equipment manufacturers will face is to achieve the timeliness and extent of these scenarios.
5G is actually an inaccurate term that is widely used. It can refer to two forms of independent and non-independent. The latter uses the existing LTE core and signaling network networking. In addition, it is also divided into millimeter wave frequency band (also known as "FR2" or "high frequency band") and sub-6GHz frequency band (also known as "FR1", consisting of "low frequency band" and "mid frequency band"). The 3GPP industry standards organization is stepping up 5G standardization work and revising Rel-15; at the same time, the Rel-16/17 standard will focus on other aspects of 5G and is expected to be approved before the end of 2022.
In addition to constantly improving technical standards, everyone is generally concerned about the 5G business model. How do operators distinguish between 5G and LTE networks? Will the 5G network achieve all or a small part of the vision?
5G network Sub-6GHz frequency band
Deploying a new generation of wireless networks is an expensive project, so operators are working hard to develop and monetize 5G applications. Although everyone has invested a lot of research and development efforts on the three major scenarios of the 5G vision, the early stage of 5G marketing mainly focused on enhancing mobile broadband (eMBB). Operators compete with each other on network coverage and speed, which also indirectly affects the Sub-6GHz network architecture and technology.
Disadvantage
If you want to compare speed or capacity, then the 5G network Sub-6GHz frequency band will not be dominant at once. This is an incidental consequence of Shannon-Hartley's law. This law describes the theoretical value of the maximum data rate that can be transmitted in a specific channel bandwidth:
C = B*log2 (1+SNR)
Among them, C is the limit of channel capacity (bit/s), B is the channel bandwidth (Hz), and SNR is the signal-to-noise ratio.
Although new Sub-6GHz frequency bands are allocated every day around the world, the bandwidth of these frequency bands can only be measured in tens or hundreds of MHz. In the millimeter wave band, the bandwidth is usually GHz level. Compared with millimeter wave, this is the fundamental disadvantage of Sub-6GHz network. Figure 2 shows how Ericsson believes that the existing LTE network should evolve to 5G with the best coverage, capacity and performance. The hybrid network combines existing 2G/3G/4G standards and frequency bands, as well as 5G Sub-6GHz and millimeter wave bands. The entire evolution process begins with carrier aggregation (CA) in different LTE frequency bands. The evolved network has dual connectivity (DC), in which the downlink operates on the 5G Sub-6GHz frequency band covering more channel bandwidth, while the uplink signal remains in the LTE network. In the end, the network is upgraded to a model that includes multiple combinations of CA and DC on the Sub-6GHz and millimeter wave frequency bands.
Advantage
The ideal situation for operators to upgrade their LTE network to a fully functional 5G network. This evolution process involves multiple frequency bands and standards, as well as CA and DC, resulting in complex and expensive implementation. Although the Sub-6GHz part of the network has the problem of insufficient channel bandwidth and increases the complexity of the hybrid network, it also brings many benefits to the 5G network.
A major advantage of the low frequency band is the signal propagation characteristics. The path loss of the transmitted signal increases with the increase of frequency in a multiple relationship of 20log10(f). In the case of the same distance, the signal loss of 28GHz is 32dB more than that of 700MHz. In view of the constant maximum transmit power of the base station, the increased path loss under this high frequency band greatly limits the coverage of 28GHz equipment. Also, the Sub-6GHz signal has lower building penetration loss than the millimeter wave signal. This is essential for the deployment of 5G networks in metropolitan areas.
Sub-6GHz networks also have obvious advantages in the application of multiple-input multiple-output (MIMO) technology and massive MIMO antennas. MIMO relies on multiple transmitters and receivers in base stations and user terminals. Because the radiators are separate, the transmitted signal arrives at the receiver along different paths. Using space diversity and multiplexing technology, coupled with single-channel multiple data streams and multi-channel propagation, can improve signal robustness (signal-to-noise ratio) and data rate.
This MIMO antenna architecture will become the mainstay of most 5G networks, because if the channel capacity in Equation 1 is approximated by a first order, the MIMO antenna can increase it by n times (n is equal to the number of antenna radiator pairs). In the standard version previously released by 3GPP, the antenna structure is limited to the 8T/8R configuration, that is, 8 transmitters and 8 receivers. The term "massive MIMO" (mMIMO) is also very general, but now it basically means that the number of transmitters far exceeds 8. In the current 5G deployment, we can see that each antenna of the mMIMO base station and access point has up to 1024 radiators.
The implementation of mMIMO on Sub-6GHz and millimeter waves PCB is different, resulting in subtle differences in architecture and design criteria. Sub-6GHz signals have longer wavelengths than millimeter wave signals, so more transmission reflections will occur. This can create a richer multipath propagation environment and give play to the advantages of MIMO. In addition, building and maintaining an optimal wireless link requires understanding channel state information, which includes processing and updating parameter information such as scattering, fading, path loss, and blocking. The above operations are more repeatable in the Sub-6GHz frequency band, thereby providing a more favorable environment for signal propagation.